Vous êtes sur la page 1sur 172

AD-A243 687

tI 11
IIlilIUN, 1 DE
AFIT/GCS/ENG/91D-17 DEC 2 7 1991
C

Evaluation of Scalar Value Estimation


Techniques For 3D Medical Imaging

THESIS
Rob W. Parrott
Captain, USAF

AFIT/GCS/ENG/91D-17

Approved for public release; distribution unlimited

91-19010
II I~IIIIIIII 91 12 24 0
Form Approved
REPORT DOCUMENTATION PAGE OMB No. 0704-0188
Public reporting ouroen for this clIlection or nformation is estimated to average I hour per respoonse. including the time for reviewing instructions. searcning existing data sources
gathering and maintaining the data needed, and completing and reviewing the ollection of information Send comments regarcing thi burden estimate or any other aspect of this
collection oftinforaaOn, nclud ng suggestions for reducing thus Ourden. :o Washington HeadQuarters Services. Directorate for information Operations and Reoorts, 1215 Jeffers.n
Davis Highway, Suite 1204, Arfirgton, i A 12202-4302, and to the Office of Management and Budget. Paperwork Reduction Project (0704.0188), Washington, DC 20503

1. AGENCY USE ONLY (Leave blank) I2. REPORT DATE 3. REPORT TYPE AND DATES COVERED
. .-... ....... ,December 1991 -.-- Master-s-Thesis-.. ... ......-

Evaluation of Scalar Value Estimation


Techniques For 3D Medical Imaging

6. AUTHOR(S)
Rob W. Parrott, Capt, USAF

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION


REPORT NUMBER
Air Force Institute of Technology, WPAFB OH 45433-6583 AFIT/GCS/ENG/91D-17

9. SPONSORINGi MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORINGi MONITORING


AGENCY REPORT NUMBER

11. SUPPLEMENTARY NOTES

12a. DISTRIBUTION/ AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE


Approved for public release; distribution unlimited

13. ABSTRACT (Maximum 200 words)


Scalar value estimation in 3D medical imaging increases data resolution for enhanced renditions and corrects
inaccurate surface formations. Accurate estimations are vital because clinical assessment is often aided by
examination of 3D medical images. This thesis explores different estimation techniques and introduces the geo-
statistical estimation technique called kriging to the field of 3D imaging. Kriging theory claims to be the optimal
estimator - better than the standard deterministic methods commonly used. The techniques investigated are
linear interpolation, trilinear interpolation, tricubic interpolation, and kriging. This research investigates scalar
value estimation in the volume pre-processing operation of slice interpolktion and in a surface extraction method
called cell subdivision. Tricubic interpolation is shown to be most useful in artificially created volumes of
smooth functions. It is also shown to produce poor results in medical volumes and in slice interpolation. More
importantly, this research demonstrates that kriging subsumes the deterministic methods investigated and can
estimate much better than tricubic interpolation.

14. SUBJECT TERMS 15. NUMBER Ur 'AGES

-C OMPUTER GRAPHICS, MEDICAL COMPUTER APPLICATIONS, INTERPO- ,17


'k;i LATION, STATISTICS,MEDICAL IMAGING, KRIGING 16. PRICE CODE
17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19.\ SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT
OF REPORT OF THIS PAGE \OF ABSTRACT
UNCLASSIFIED I UNCLASSIFIED IUNCLASSIFIED UL
NSN 7540-01-280-5500 Standard Form 298 (Rev 2-89)
Prescribed by ANSI Std Z39-18
28-102
AFIT/GCS/ENG/91D-17

Evaluation of Scalar Value Estimation Techniques For 3D Medical


Imaging

THESIS

Presented to the Faculty of the School of Engineering


of the Air Force Institute of Technology
Air University

In Partial Fulfillment of theA


Requirements for the Degree of : LL*' ?*

Master of Science in Computer Systems ,

F.,-

Rob W. Parrott, B.S. in Computer Science '.


Captain. USAF /
December. 1991

Approved for public release; distribution unlimited


Table of Contents

Page

Table of Contents........ .. .. .. .. .. .. .. .. .. .. .. .. .. .....

ist of Figures .. .. .. .. ... ... ... ... ... ... .. ... .... vii

List of Tfables. .. .. .. .. ... ... .. ... ... ... ... ... .... xi

Acknowledgments................ . . ....
.. .....
. .... .... xi

Ab~stract .. .. .. .. .. ... ... .. ... ... ... ... ... ....... xiii

1. Introduction.. . ... .. .. .. .. ... ... ... .. ... ...... 1-1


1.1 Background. .. . .... ... ... ... .. ....... 1-2
1.2 Problem. .. .. .. .. ... .. ... ... ... ...... 1-5
1.3 Purpose. . . .. ... ... ... ... ... .. ...... 1-7
1.4 Approach .. .. .. .. ... ... ... ... .. ...... 1-S
1.5 Overview .. .. .. .. ... ..... ... ... ..... 1-9

11. 3D Medical Imaging .. .. .. .. ... ... ... ... ... ..... 2-1
2.1 3D) Data Representation .. .. .. .. ... ... ... 9_-9

2.2 3D Medical Imaging Transformations .. .. .. .. ..... 2-3


2.3 3D) Medical Imaging Methods .. .. .. ......... 2-3
2.3.1 Contour Methods .. .. .. .. .. ... ...... 2-6
2.3.2 \'oltime Methods .. .. .. .. ... ... ..... 2-6
2.3.3 Surface Methods. .. .. .. ... ... .. .... 2-7
2.4 Chapter Summary. .. .. .. ... .. ... ... ..... 2-2-3

ii
Page
III. Kriging Theory 3-1
3. a kg o . . . . . . . . . . ., . . ., ., . . ,. . . . . . 3-1
3.1 Background............................... 3-1

3.2 Kriging and Least Squares . ............... 3-4


3.3 Derivation of the Ordinary, Point Kriging Equations. . 3-4
3.4 Kriging Categories and Types ............... 3-6

3.4.1 Kriging Categories ..... ............... 3-7


3.4.2 Kriging Types ........................ 3-7
3.5 Structural Analysis ...... .................... 3-13
3.6 Isotropy/Anisotropy ...... ................... 3-20
3.7 Chapter Summary ...... .................... 3-20

IV. Cell Subdivision and Slice Interpolation Implementation 4-1


4.1 Introduction ................................ 4-1
4.2 Overview of Cell Subdivision and 3D Surface Formation 4-1
4.3 Tricubic Interpolation ..................... 4-7
4.4 Kriging Estimation ...................... 4-9

4.4.1 Overview of Kriging Technique ......... ..... 4-9


4.4.2 Global and Local Drift ................. 4-10
4.4.3 The Assumed Model Semivariogram ...... 4-11
4.4.4 Neighborhood Size ............... ..... 4-11
4.4.5 Estimation Procedure .................... 4-13
4.5 Slice Interpolation. .......................... 4-15

V. Results ........ .................................. 5-1


5.1 Artificial volume. ....................... 5-2
5.1.1 Neighborhood size 64, subdivision factor = 4.
local drift assumption differs ............... 5-3
5.1.2 Neighborhood size 61. subdivision factors differ.
no local drift assumed . ............ 5-3

iii
Page
5.1.3 Neighborhood size differs, Subdivision factor =
2, local drift assumption differs ........ 5-6
5.2 Medical image slice interpolation ................ 5-12

5.2.1 Dog heart, CT....................... 5-12


5.2.2 Baby head, MRI ...................... 5-19

5.3 Medical volume cell interpolation surface extraction . . 5-19


5.4 Chapter summary .......................... 5-22

VI. Recommendations and Conclusion ........................ 6-1

6.1 3D Imaging Recommendations .................... 6-1


6.2 Kriging Recommendations .... ............... 6-2
6.3 Conclusion ................................ 6-3

Appendix A. Vanilla Marching Cubes Data Flow Diagrams and Pro-


gram Description ........................... A-1
A.1 Data Flow Diagrams ...... ................... A-I
A.2 Program Description Language Statements ........ ... A-I
A.2.1
parse command line ("Top level" DFI), newline
Bubble 1)

A-6
A.2.2
read header info ("Top Level" DFD. Bubble 2)

A-7
A.2.3
allocate memory storage ("Top LevelF )FD, Bub-
ble 3)

A-8

i'"
Page
A.2.4
process slices ("Top Level" DFD, Bubble ,,

A-8
A.2.5
march between slices ('"process slices" DFD, Bub-
ble 4.6)

A-10
A.2.6
write box ("Top Level~ DFD, Bubble 5)

A-12
A.2.7
output geometry file ("Top Level" DFD, Bubble
6)

A-12
A.3 Data Dictionary For Data Flow Diagri ms ........ .... A-12

Appendix B., Vanilla Marching Cubes ................. B-i

B.1 Introduction ............................... B-1


B.2 Background ...... .......................... B-2
B.3 The Main Steps in the Implementation ....... . .. B-5

B.4 Precalculated Table .......................... B-7

B.5 Cell Edge Interpolation ....................... B-9


B.6 Normal Calculations ..... .................... B-12

B.7 Fixes to Pubiic Domain Code ................... B-13


B.8 Marching Cubes Output ..................... B-15

V
Page
Appendix C. Cell Subdivision Implementation. .. .. .. .. ...... C-1

C.1 Terminology .. .. .. .. .. ... ... ... .. ...... C-i


C.2 Purposes of Cell Subdivision .. .. .. .. .. ... ..... C-i
C.3 Implementation Steps. .. .. .. ... ... ... ..... C-3

Appendix D. Disambiguation. and Enhanced Surface Representation


by Cell Subdivision .. .. .. .. .. ... .. ... .. D-1
D.1 Background. .. .. .. .. ... ... ... ... ...... D-1
D.2 Example . ..... .. .. .. .. .. .. .. .. . .D-2

Appendix E. Binary Image Format to Utah RLE Format Conversion E-1

Appendix F. Changes to the Air Force Institute of Technology's Gen-


eral Purpose Renderer .. .. .. .. .. ... ... ..... F-1

Appendix G. Creating Artificial Volumes. .. .. .. .. ... ...... G-1

Appendix H. Trilinear Interpolation .. .. .. .. ... .. ... .... H-1

Bibliography. . ... .. .. .. .. ... ... ... ... ... ... ...... BIB- I

Vita .. .. .. .. .. ... ... ... ... ... .. ... ... ... .... VITA-i

vi1
List of Figures
Figure Page

2.1., Pictorial representation of a medical data volume .......... 2-2

2.2. Pictorial Representation of 3D Medical Imaging Transformations 2-4


2.3. Classification of 3D Medical Imaging Methods ............ .... 2-5
2.4. Computational Cell ..... ......................... 2-10
2.5. Ambiguous face - intersection point connection choices ...... .... 2-12
2.6. First 5 cases of 15 cube vertex classifications and triangulations 2-14
2.7. 5 more cases of 15 cube vertex classifications and triangulations 2-15
2.8. Final 5 cases of 15 cube vertex classifications and triangulations 2-16
2.9. Facial averaging example .............................. 2-17

3.1. Kriging Example ................................... 3-2


3.2. Table of Kriging Types (All three can estimate points or blocks). 3-8
3.3. Global and Local Drift ............................... 3-11
3.4. Example semivariograms for linear, spherical and exponentional
models ........ ................................ 3-18

4.1. Computational cell (cube) marching between data slices ..... 4-3
1.2. Subdivided Major Cell ..................... -1-4
4.3. Major cell centered within surrounding cube ................ 4-5
4.4. Example of fitting points to a curved line in 2D .............. 4-7
.1.5. Drift model %ariogram ....... ........................ 4-12

.5.1. Vanilla marching cubes extraction of hyperboloid surface ..... .. .5-4


5.2. Cell subdivisionfactor 2. with trilinear interpolation est imating minor-
voxel values and marching cubes extraction of hyperboloid surface
from mini-slices ........ .............................. 5-6

vii
Figure Page
5.3. Cell subdivision, factor 2, with t :cubic interpolation estimating
minor-voxel values and marching cubes extraction of hyperboloid
surface from mini-slices ....... ....................... 5-7
5.4. Cell subdivision, factor 2. with kriging estimating minor-voxel val-
ues and marching cubes extraction of hyperboloid surface from mini-
slices. Kriging uses a neighborhood of 64 sample values and assumes
no local drift .................................... 5-8
5.5. Subdivision factor 3. Upper left, vanilla marching cubes. Up-
per right, trilinear interpolation, Lower left, tricubic interpolation.
Lower right, kriging with 64 neighborhood, no drift .......... 5-10
5.6. Subdivision factor 4. Upper left, vanilla marching cubes. Up-
per right, trilinear interpolation. Lower left, tricubic interpolation.
Lower right, kriging with 64 neighborhood, no drift ........... 5-11
5.7. Subdivision factor 5. Upper left, vanilla marching cubes. Up-
per right, trilinear interpolation. Lower left, tricubic interpolation.
Lower right, kriging with 64 neighborhood, no drift ........ .... 5-12
.5.8. Kriging neighborhoods, nh32, nhl6x, nhl6y and nhl6z ...... .... 5-13
5.9. Kriging estimation, subdivision factor 2, neighborhood 32. Upper
left, no drift. Upper right., local linear drift. Lower left, image dif-
ference of upper right from upper left. Lower right., image difference
of upper right image from f2tricubic2 ........... I....... 5-14
5.10. Kriging estimation, subdivision factor 2, neighborhood 16x. Upper
left, no drift. Upper right, local linear drift. Lower left. image dif-
ference of upper right from upper left. Lower right, image difference
of upper right image from f2trilinear2. . .. .......... 5-15
5.11. Kriging estimation. subdivision factor 2. neighborhood 16y. Upper
left. no drift. l pper right, local linear drift. Lower left., image dif-
ference of upper right from tipper left. Lower right, image difference
of upper right image from f2tricubic2. .................. 5-16
5.12. Kriging estimation, subdivision factor 2, neighborhood 16z. Upper
left, no drift.. Upper right, local linear drift. Lower left, image dif-
ference of upper right from upper left. Lower right, image difference
of upper right image from f2trilinear2 ..................... 5-17

viii
Figure Page
5.13. Kriging estimation, subdivision factor 2, neighborhood 8. Upper
left, no drift. Upper right, local linear drift. Lower left, image dif-
ference of upper right from upper left,. Lower right, image difference
of upper right image from f2trilinear2 ..................... 5-18
5.14. Dog heart CT medical image slice interpolat.ion. Create new slice
between slices 41 and 42. Window titles depict type of estimation
performed ....... ............................... 5-22
5.15. Baby head, MRI medical image slice interpolation. Create new slice
between slices 32 and 33. Window titles depict type of estimation
performed ....... ............................... 5-23
5.16. Baby skin,iso-value 43, Vanilla Marching Cubes ............. 5-23
5.17. Baby skin,iso-value 43, Cell Subdivision, Trilinear Interpolation 5-24
5.18. Baby skin,iso-value 43, Cell Subdivision, Tricubic Interpolation 5-24
5.19. Baby skin,iso-value 43, Cell Subdivision, Kriging ... ......... 5-25
5.20. Baby skin,Difference image between images produced by trilinear
interpolation and kriging ....... ...................... 5-25

A.l. Vanilla Marching Cubes Context Diagram ................. A-2


A.2. Vanilla Marching Cubes Top Level DFD. ................... A-3
A.3. Vanilla Marching Cubes "process slices" DFD ... .... . A-4
A.4. Vanilla Marching Cubes "march between slices" DFD ....... .... A-5

B.1. Computational cell (cube) marching between data slices ..... B-6
B.2. Example of cell mapping to a unique case ................. B-8
B.3. Correspondence of interpolation arravs to computational cells marcli-
ing between two slices I..................... B-II
B.4. Vanilla marching cubes example boundary cases in surface normal
estimation ....................................... B-14

C.1. Subdivided Major Cell ....... ........................ C-2


('.2. Computing subdivision points. Subdivision factor = 5 in all three'
directions ........................................ C-6

ix
Figure nage
D.I1. Alternate polygonization for case 6. .. .. .. .. ... ... ..... D-3
D.2. Example of alternate surface representation caused by cell subdivi-
sion .. .. .. ... ... .... ... ... ... .... ....... D-.5

11.1. Trilinear interpolation .. .. .. .. ... .. ... ... ... ..... H-2


List of Tables
Table Page

5.1. Comparison of values estimated by trilinear and tricubic interpola-


tion and kriging (with a. 64 neighborhood, no local drift) ..... 5-5
5.2. Comparison of values estimated by different kriging forms for cell
interpolation surface extraction of a hyperboloid surface in an arti-
ficial volume ........ ............................. 5-9
5.3. Comparison of dog heart CT estimated values........... . 5-20

5.4.. Comparision of baby head MRI Limated values........... 5-21

xi
Acknowledgm.ents
My first and ultimate thanks go to my Lord and Saviour, Jesus Christ. Through
Him I found the strength and confidence to keep going when I thought I wasn't ca-
pable. I thank God, His Holy Spirit, and His Son for being with me and my family.
With Them in our lives and Their eternal perspective in our hearts, we grew stronger
in our faith during this time. Thank you Lord for making this a learning and at times
an enjoyable experience for me and one that I will cherish for the rest of my life.
Special thanks go to my family. Thank you Verla for supporting me through
all the times I wasn't with you, especially weekends (I know. what's a weekend?).
I'm very grateful for your strength and patience. Misty, Alex, and Wesley - I'm
sorry you had a part-time Dad for the last 18 months, but I'm back full-time and
we're a family once again!
I would like to thank my advisor. Col Martin Stytz,, who gave me invaluable
support and guidance in the area of 3D medical imaging and technical writing. His
previous papers and dissertation greatly aided in my understanding of 3D medical
imaging. Thank you Col Stytz for your constant encouragement and positive outlook.
These qualities helped keep me motivated throughout the research.
Thanks also go to Col Phil Amburn and Maj David Robinson. who both pro-
vided me with crucial technical guidance. Thsnk you Col Ainburn for first recom-
mending the use of marching cubes and kriging in this research. Thank you also for
your help with GPR, tricubic interpolation, an(l numerous other areas where your
keen insight pushed me in the right direction. Thank you Maj Robinson for your
assistance in the area of kriging. With Your deep understanding of this complex
estimation technique, you helped me a great deal to see how it could be applied in
3D imaging.
Thanks go to Capt Chris Brodkin for sharing his kriging code with me. \Vith-
out, this. my work would have been much more difficult.
Finally. Yd like to thank the fellow students who listened to me when I needed
feedback. This act of kindness was a tremendous aid when code was seemingly
debug-proof or concepts just didn't seem clear. Thanks for your comradeship - Pat
Rizzuto. Patiy Brightbill. Don Duckett. and Wayne Mcgee.

Rol) NV. Parrott

Nii
AFiT/GCS/ENG/91 D-17

Abstract
Estimation has not received enough attention in 3D medical imaging. Estima-
tion is often done in 3D medical imaging to increase data resolution for enhanced
renditions. It is also used for correcting inaccurate surface formations in the well
known marching cubes algorithm. Accurate estimations are vital because clinical as-
sessment is often aided by examination of 3D medical images. This thesis 'introduces
the geo-statistical estimation technique called kriging to the field of 3D imaging.
Kriging theory claims to be the optimal estimator - better than the standard deter-
ministic methods commonly used.
This thesis explores four estimation techniques for use in 3D medical imaging.
The techniques are linear interpolation, trilinear interpolation, tricubic interpolation,
and kriging. The interpolation methods are standard estimation techniques used
in 3D imaging. The estimation techniques are used to estimate scalar values in
two primary areas. These are intra-cell scalar value estimation and the volume
pre-processing operation of slice interpolation. This research investigates intra-cell
scalar value estimation in a surface extraction method called cell subdivision. This
research also explores slice interpolation by estimating scalar values between existing
medical data slices. Slice interpolation is the operation of estimating logical slices
between existing ones, typically to increase data. resolution to obtain a finer mesh
representation of a surface.
Tricubic interpolation is shown to be most useful in artificially created volumes
of smooth functions. It, is also shown to produce poor results in medical volumes
and in slice interpolation. More importantly, this research demonstrates that kriging
subsumes the deterministic methods investigated and can estimate much better than
tricubic interpolation.

Xii,
Evaluation of Scalar Value Estimation Techniques For 3D Medical
Imaging

I. Introduction

A twelve-year-old boy suffers from severe congenital defects in the sacrum,


lumbar spine, and acetabula (hip). The surgeon assigned to this case reviews both
two-dimensional (2D) and three-dimensional (3D) medical images of the boys pelvic
region. The 3D image is generated by a computer graphics program with input from
a series of 2D Computed Tomography slices (scans) of the patient's pelvic region.
3D medical images provide views of internal organs, bones, tissues, muscles and
other body parts otherwise only seen after the human body has been invaded by
exploratory or therapeutic surgery. Because of the additional insight revealed by the
3D images, the surgeon decides not to perform surgery (45). In this case, 2D scans
alone do not provide enough information for the surgeon to make an accurate as-
sessment of the patient's condition. Analysis of craniofacial abnormalities, radiation
treatment planning for cancer patients. analysis of pelvic deformities , and cardio-
pulmonary analysis are only a few of the areas 3D medical images have assisted
medical clinicians make vital decisions in patient management (45).
Because of the need to visualize volume (3D) data sets for analysis, computer
scientists have developed many different techniques for portraying this data. Met h-
ods exist to depict this three-dimensional data in 3D; however, the emlphasis has
been to display the 3D data on a 2D screen (26) and (27). This research explores
a subset of the latter methods in the area of niedical imaging, with an emphasis on
estimation of scalar values.,
To understand the exact problem considered and the purpose of this thesis
-me background information must be presented first. Following the background,
the problem is discussed. This chapter ends by presenting the purpose of the thesis
followed by the research goals.

1-I
1. Background
Three dimensional data visualization encompasses all methods that render im-
ages from volume data sets. Medical imaging and scientific visualization are two
broad application areas under 3D data visualization. The need for accurate rep-
resentations of the physical volume separates medical imaging and many scientific
visualization areas from other 3D computer graphics techniques. The latter tech-
niques mostly attempt "... to form realistic images from scene descriptions which
may, or may not, have a physical counterpart" (41:2). Accuracy in medical images
is important because many clinical applications rely on the faithful representation
of the portions of the human body scanned and reproduced three-dimensionally.

3D medical imaging is composed of three high level processes : "&.ta collection,


3D data display, and data analysis" (41:2). This work is concerned with 3D data
display - a computer graphics task. Data collection and data analysis are both
performed by end-user experts - radiologists and clinicians.
Data collection is defined by Stytz (41:2) as

...the radiologist's arena, and deals with issues concerning the statistical
significance of collected data. patient dosage, medical imaging modality
operation, development of new techniiques/modal ities for gathering data,
and 2D image reconstruction.

The main data, collection scanning technologies (or modalities) include Computed
Tomography (CT), Magnetic Resonance Imaging (MRI). Single Photon Emission
Computerized Tomography (SPECT). and Positron Emission Tomography (PET)
(41) and (39). The main purpose of each of these modalities is to sample data in
patient space and produce a series of 2D images corresponding to some aspect of
the patient such as soft and bony tissue in the case of CT andI MRI studies and
metabolic measurements in the case of PET and SPECT studies. CT and MRI
are used primarily for imaging anatomical structures. whereas SPECT and PET are
used mainly for biochemical imaging. For each modality. different properties of the
human body are imaged, but all produce a series of gray scale 2D images. Gray scale
means values range from 0 to 255 (8 bits per value). For the purposes of this thesis,
it is enough to say the values correspond to some material prol)erty" of the imaged
volume, such as density.

1-2
CT was the first modality, folloNN ing the ad-,t of X-ray technology, to provide
high quality 2D non-invasive images. The others tollowed soon thereafter to provide
images of equal (if not. better) quality. Since this research deals with 3D data display,
reference Stytz for aa overview of how each of these modalities are constructed and
how their operation influences image quality. (39).
These 2D images alone provide a certain level of non-invasive assistance, but
with these images the clinician must mentally build 3D objects of interest. This may
be very difficult, depending on the experience of the analyzer and the application
area. In one area of clinical study, according to Wojcik and Harris (45:197),

Mental integration of all of the images frequently obtained to evaluate


a complex skeletal problem (i.e., plain radiographs, polydirectional to-
mography, and MPCT [multiplanar CT]) is freqtcrdy difficult even for
experienced radiologists and is even more difficult for less experienced
radiologists or nonradiologists [such as surgeons].

A four year study that assessed 3D imaging from CT scans at. the Department of
Diagnostic Radiology, University of Manchester concluded that "3D imaging does
have a useful role to play in a number of specific clinical situation- when used in
conjunction with CT and other radiological imaging methods" (45:103-144).
The main challenge faced by 3D data display graphics reseairchers is how to
create clinically useful accurate 3D images quickly from the inmeise amount of in-
formation obtained from these imaging modalities. One modality can produce up to
35 mega-bytes of data for a single patient (41). This occurs becav, e a typical study
consists of 12 to 100 or more slices (scans) of images with resolutions up to 512 X 512.
Following the development of CT in the early 1970's. many software algorithms were
created to produce 3D images from CT data. The methods conceptually combine the
2D CT data into a volume of information from which significant. data is portrayed.
Most other 3D visualization methods (such as scientists and engineers studying com-
putation fluid dynamics, molecular modelling oi lhe eart h sciences (:36)) have their
roots in early medical imaging techniques (45:60). These methods generally fall into
two main categories - volume and surface methods.
Most 3D imaging researchers use the terms volume rendering and surface ren-
dering to classify rendering of volume (3D) data sets. Surface rendering methods

I-3
extract a surface or surface boundary from tie volume data set and represent it using
some data siructure. Volume rendering directly processes voxels (volume elements),
assuming that each voxel is either opaque or partially opaque. This classification
can be misleading in some cases. because it implies that surface rendering only ren-
ders surfaces and volume renderers only render volumes. Yet, surface renderers can
render volumes (multi-surfaces using opaque and transparent surfaces) as well as
just single surfaces. Also, volume renderers can display distinct surfaces much like
surface renderers by processing binary volumes (45). A binary volume is formed by
processing only those voxe. forming part of a surface of interest. Then only the
the selected voxels are rendered. It is termed binary because voxels are classified as
contributing to a structure of interest (assigned a 1) or not (assigned a 0).
I prefe," Farrell's terminology (19) of surface unit based and volume mnit based
approaches, that differentiates not by what the final image consists of, but rather by
the type of unit or "primitive image element" used in the visualization process. For
example, two-dimensional primitive image units, such as planar polygons or voxel
faces, are used in the surface unit based appr jaches, whereas volume elements (vox-
els) is one type of three-dimensional primitiv- image un., used in volume unit based
approaches (see chapter two for a more con plete discussion of these approaches).
Hereafter in this document. the terminology , urface method and volume method cor-
responds to surfa, e unit based and volume unit based approaches. respectively.
Surface and volume methods have their supporters in both the scientific and
medical communities. The choice of method depends on the application and often on
the personal preference of the researcher. scientist, eoigineer, radiologist. or clinician.
Many researchers (.e.g (47),(43), and (27)) have noted that a variety of approache,
used to render the same data set provide crucial clues to analyzing and unders".aiid-
ing the data, whereas only one approd. 1 might leave gaps in knowledge. In the case
of medical imaging. surface methods are good f-;, :i.wplaying surfaces with definite
boundaries. such a.s bony tissue in the human body. Likewise, volume approaches
are superior to surface approaches for depir ,ag fuzzy or amorphous volumes such
as diffused tumors and blood flow in arteries and veins (33), (16) and (45). Also,
surface methods allow for one of the most useful medical imaging applications -
interacgive manipulation of structures (45). One purpose of this task is surgery re-
hearsal. Surgery rehearsai is not very practical with images rendered via volume
methods because they typically are too slow for interaction. Surface methods allow

1-4
real-time manipulation because thy normally produce. geometric primitives such is
planar polygons. These geometric primitives ca. be used as input into fast hardware
or software based hidden surface removal and shading algorithms. However, if a new
surface is desired for rendering, the entire volume must be re-processed again. Vol-
ume methods require a ,igh computatir-, .' -,st for processing all the data elements
in the volume; however, the increase,'( ^", -,aybe less important than the need for
a diffused or fuzzy rendition of th.- . .a.

The third medical imaging proce:'-.- data analysis, involves obtaining "quan-
titative information about the structur . ,ntb:, scene" (45:57) Quantitative infor-
mation includes the average density of " Earea v ;thin the 3D image, the size of
certain anatomical structure. uct as bone.; and biaod vessels, or the volume of a
sub-region. Much of the desired analytic information can be acquired directly fror'l
the 2D slices. Measurements reqL.: -d from 3D images are volumes, 3D distances, 3D
angle measurements, and "other less commonly used measurements... [such as] cen-
ter of mass, moment of inertia, and surface (urvature." Surface methods as well as
certain volume methods (which allow the concept of a structure) provide the ability
to perform all the 3D data analysis operations (45:58).
Image accuracy is critical to the data znalysis process. Inaccu'acies in an image
can adversely affect these measurementc. possibly leading to erroneous assessments
by the chinician.

1.2 Problem
Scalar value estimation is very common inmany 3D imaging algorithms; how-
ever, little has been done to investigate different scalar value estimation teclniques.
The level of inaccuracy has not. been deemed serious b'.cause cost. has been a
larger concern. As Herman and Liu noted in 1979, "... reduction in cost is essential;
computer time for the display will bave to be borne by the patient." Cost is certainly
still an issue (45:224). Higher order deterministic functions such as quadratic or cu-
bic polynomials and statistical based estimation methods are more computationally
expensive than simple linear interpolation. Yet. the results obtained from using them
can greatly improve image accuracy and fidelity - thus improing clinical assessment
and quantitative analysis. Since work itations are becoming more p.rl, this issue
of cost will become less of a problem; hence. a search for more accurate estimation
meto(s is necessary.

1-5
Accuracy in medical imaging is very important to the medical community.
Wojcik and Hari; note, "the primary purpose of radiolcgists is to provide the most
accurate diagnostic information possible within the capabilities of available imaging
mrdalities" (45:196). This estimation problem inay cause serious 3D image errors in
two primary ares:

e The volume pre-processing operation of slice interpolation.

• Intra-cell scalar value estimation1 .

.idupa and Herman s~ate, "the interpolation problem in our opinion, ias
received less attention than it deserves" (45:13). The interpol ion problem they are
referring to is the process of J,.tcrmining new slices between existing image slices to
form a cube shaped volume. Udupa and Herman discuss the primary methods of
interpolating new slices are nea,'est neighbor,linear interpolation of voxel values, and
trilinear interpolation. The nearest iieighbor approach produces the worst results of
the three because it does no estimation. This approach simply assigns new voxel
values to the iuerpolated slices to the nearest original voxel. Trilinear performs
somewhat better than linear, but the variation in both cases is still assumed to
be linear when in fact it may not be. Udupa and Herman investigated one other
method they created for use i 'amary volumes. This method is termed shape based
interpolation. Voxels ai-e mapped from the binary volume to a seconm: , orrespording
array that holds distances from each voxel to the boundary. If the voxel in ti,c boinary
volume was 1, the distance in the second corresponding array will be poitive, else
negative. Linear, trilinear, or some other interpolation scheme is used to derive new
d&tances in the second array. The newly interpolated distances are then mapped to a
ne"," binary volume consisting of new slices betwetia existing ones. Positive distances
are it:,Signed! a 1 in the binary volume, negative distances a 0. Hence. the boundary
of the structure of interest b)rtwee . i9 and 1 entries in the binary volume influences
the int.erpola'ion. The authors claim it. provides more accurate quantitative analysis
and in their opinion leads to a better surface representation in surface methods.
However, this method will not work in volume methods unless the volume melod
uses a binary volume.

IA cell is a logical cube with vertices formed by four voxels in one data slice and four voxels in
an adjacent slice.

1-6
Many 3D rendering algorithms (primarily surface methods) require cubic shaped
voxels (28), (44), (22), and (45). Voxels are volume elements, the 3D analog of pix-
els. Cube shaped voxels are termed cuberilles. Other rendering algorithms, including
volume methods such as (33:34) , interpolate new slices to improve image quality.
For example, assume a study consists of a series of 100 256 pixel X 256 pixel CT
images. To make this volume cubic in shape, new slices are created in between exist-
ing ones until there are 256 total sli,, s. The extra 156 slices are usually determined
by linear interpolation of the sampled CT values between the original shces. The
problem with this method is the data may not vary linearly.
Intra-cell scalar values are estimated to improve imagP fidelity and correct pos-
sibly inaccurate surface renditions generatd.l by cell interpolation surface methods2 .
Wilhelms and Gelder (50) have shown that the commonly used trilinear interpolation
estimation method does not etimate values in artificially created volumes (versus
scanner generated) as accurately as - parametric cubic function. They demonstrated
that tricubic interpolation prod'ices better images than those produced by trilinear
interpolation.
The problems with linear, trilinear, and tricubic interpolations are that these
methods assume the variability of the data and assT a-. a neighborhood of sample
values that influence the estimation. However, variation is not necessarily linear or
cubic in nature and the number of saml)le values that .hould influence the estimation
can be different than the unchangeable number determined by these methods. As-
sumptions about the variation of the data and the ncghlbohood of sample values can
possibly produce erroneous results in the estimation process. Instead of assuming a
variation, a geo-statistical process exists that can aid in determining the variation of
the data for the purpose of estimating new values. In thib same process, the number
of sample values influencing the estimation can be modified to fit. the variability of
the data. This process is termed kriging.

1.3 Purpose
The purpose of this research is to investigate the application of several esti-
wation techniques to estimating scalar values within co-nputational cells and during
tWe volume pre-processing operation of slice interpolation.

2
Cell interpolation surface methods are described in chapter two.

I-7
To accomplish the purpose of this thesis, four estimation techniques were irn-
plemented. These are linear interpolation, trilinear interpolation, tricubic interpo-
lation, and kriging. This is the first time the geo-statistical estimation technique
called kriging has been applied in 3D medical imaging.

Kriging differs from the deterministic methods implemented 3 because it is a


statistical method. That is, kriging accounts for error Not only is it a statistical
estimation multiple linear regression model, but it is a geo-statistical model as well.
A geo-statistic differs from a classical statistic by how the variables are treated in
the estimation process. A classical statistic views data as random. A geo-statistic
accounts for spatial inter-dependence of the variables used in the estimation. Kriging
has proven to be very successful in estimation applications in the geo-sciences. Since
medical data is spatially distributed and has regions in which sample values are highly
correlated (e.g., bone and tissue regions), this technique is applicable to medical data.

Kriging theory shows that kriging is the optimal estimator. It is optimal in a


statistical sense in that it minimizes estimation error variance and removes bias in
the estimation. This research demonstrates that this optimal estimator subsumes
the deterministic methods investigated and in some cases produces visually better
results.

1.4 Approach

The main goal of this research is to compare the results of different techniques
for estimating scalar values. Estimated values are used in two areas - within compu-
tational cells and for creating logical slices during slice interpolation. To accomplish
this goal, the following tasks were accomplished :

" A cell interpolation algorithm was implemented, including the capability of


subdividing cells for int.ra-cell scalar value estimation.

" Trilinear interpolation, tricubic interpolation. and kriging estimation functions


were implemented for comparison.

" Artificial volume data sets were built for comparing and contrasting methods.

3 The deterministic methods are linear. trilinear. and tricubic interpolation. Determinism means
t-hey do not account for any variation such as systematic error in sampling.

1-8
* A rendering system was developed by modifying an existing one (to support
software engineering goal of re-usability) to display images for comparison.
* 3D Images were generated by extracting surfaces from both artificial volumes
and actual medical data sets by using trilinear interpolation, tricubic interpo-
lation and kriging estimation techniques.
* 2D Images were generated by estimating scalar values between two medical
image slices using linear interpolation, tricubic interpolation, and kriging esti-
mation techniques.
* Values and images derived from the different estimation techniques were com-
pared.

1.5 Ovemiew

The remainder of this document consists of the following chapters. Chapter


two provides an introduction to 3D medical imaging methods: focusing mainly on
surface methods., Chapter three describes kriving theory. Chapter four provides
implementation details of the algorithms developed during this effort. Chapter five
contains the images resulting from the research and a discussion of these results.
Chapter six consists of a section recommending future research following this effort
an(l a section with concluding remarks. The appendices provide further implemen-
tation details for the interested reader.

1-9
II. 3D Medical Imaging

This chapter discusses 3D medical imaging concepts, concentrating on the areas


most pertinent to this research. In particular, 3D data representation and 3D medical
imaging transformations are summarized, followed by a discussion of the major 3D
medical imaging methods - contour, surface and volume. Because they were used in
this research, surface methods are discussed in more detail than the other two areas.
More comprehensive backgrounds of 3D medical imaging concepts can be found in
(45), (19), and (41).
I chose to implement a cell interpolation surface method - based on the march-
ing cubes algorithm (34) (described later in this chapter) - because the subdivision
technique developed by Wilhelms and Gelder (50) provided me with an excellent
framework to explore kriging in the intra-cell estimation process.
Inaccuracies occur in the marching cubes algorithm because it can incorrectly
define the surface of interest in a volume. The algorithm assumes the original sam-
pled data is suff,:ient to represent the surface accurately, when in fact it may not
be. A correct rei. -lering of the surface never occurs because surfaces are continuous
and we are sampling from this continuous domain. The best alternative would be
to sample finer, but with current scanning technology limitations. costs, and pa-
tient health, this is not always possible. Because obtaining a finer sampling can be
difficult and costly, Lorensen and Cline (8) and Wilhelms and Gelder (50) obtain
more accurate renditions by subdividing portions of the volume and estimating new
scalar values within the subdivided volume areas, given a surrounding neighborhood
of original sampled values.
Wilhelms and Gelker (50) investigated two estimation methods -- trilinear and
tricubic interpolation. In this thesis effort. I explore a statistical estimation func-
tion derived by a process called kriging. hriging, according to Matheron. is the
probabilistic process of obtaining the best linear unbiased estimator of all unknown
variable' (25). If properly applied, kriging. in contrast to trilinear and trictibic. an-
alyzes the data prior to estimation to determine the actual variation of the data and
minimizes the estimation error variance. See chapter three for a detailed discussion
of kriging.

2-1
2.1 3D Data Reprtsentation
To understand how medical images are produced, one must first become fa-mil-
iar with some of the basic concepts of 3D imaging from 3D data sets.
Most algorithms developed for rendering 3D data sets expect the data to be
regular shaped, or in a 3D lattice or grid format. A lattice or grid in this context
is best visualized as a framework of parallel planes in space, with data information
(such as material density) conceptually located at the intersection of the planes
or within the volume of space between planes. In many cases transformations are
applied to the discretized data to remove sampling noise, to alter the resolution of
the data. or to make an irregular data set regular (45:4-14).
3D medical data is regular because it is derived from conceptually stacking 2D
same resolution scans. e.g., from CT or MRI output. The third dimension i, then
realized as the stack or slice number (see figure 2.1). Regular shaped data is popular

Figure 2.1. Pictorial represenltationl of a medical data voluin.-

because it call be vasily [napped directly into a 31) array. Once in the array. di

.) .2
is ordered for manipulation. One method of ordering is simply to view elemtnts
in the array as volume elements, or voxels. Voxels are conceptually the equal-sized
parallelepipeds created by the intersections of three sets of parallel planes, each set
orthogonal to the other two. The voxel and cuberille data models have been widely
used in imaging agorithms. The use of these data models is discussed in the section
titled 3D Medical Imaging Methods.

2.2 3D Medical Imaging Transformations


The 3D medical imaging transformations can be seen in figure figure 2.2,
adapted from (45). These are operations upon the data that transform it from
scanner output to a 3D graphics image projected onto a 2D screen. The medical
image data, input into scene space, are the 2D image slices output from scanners.
One scene transformation from scene space to scene space processes the input data
by filtering it to suppress noise or to determine tissue boundaries. Forming new
slices by estimating values between existing images is a common scene transforma-
tion (also called slice interpolation). Another scene transformation is a Volume Of
Interest (VOI) operation. A VOI operation pre-processes the scene data to include
only the data that contributes to a structure of interest. A structure of interest
might be a single organ or a. portion of a volume, like the left hemisphere of a brain
study. Structure extraction is the step in the 3D medical imaging pipeline that gen-
erates some structure, such as a polygon mesh, from scene space into object space.
A VOI operation only encloses the structure of interest in scene space whereas strut -
ture extraction generates a separate data representation of the structure of interest.:
Structures in object space are possibly transformed by geometr:- transformations
(rotations, translations, or scalings) into image space. Rendering (projective trans-
formations) transforms the image space structures into view space - the 2D computer
graphics screen. The final data transformation, analysis. acts on the 2D image to
provide parameters for analytic measurements, discussed as data analysis in chapter
one.

2.3 3D Medical Imaging Methods

3D medical imaging methods are typically divided into three main approaches:
contour, surface and volume. The techniques differ in the dimensionality of the
geometric data model used to create the image - ID contours (ID units) for the

2-3
medical image
data

SCENE scene
SPACE transformation

Structure
extraction

OBJECT ] structure
SPACE transformation
I geometric
transformation
IMAGE H geometric
SPACE transformation

I projective
transformations

[ VIEW
SPACE
image
processing

I analysis

PARAMETER
SPACE

Figure 2.2. Pictorial Representation of 3D Medical Imaging Transformations

2-4
contour approaches, 2D polygons (2D units) for the surface methods, and voxels and
computational cells (3D units) for the volume methods (See figure 2.3). Only those

3D Medical
Imaging Approaches

Contour Methods Surface Methods [Volume Methods I

(11) Units) (21) Units) Vo(31) Units)

Tiling ISurface
Tracking Cell Interpolation

Figure 2.3. Classification of 3D Medical Imaging Methods

medical imaging topics that most directly relate to my work are discussed in this
section. Rendering (projective transformations) is a major 3D imaging task, but is
not discussed at length here because this research emphasizes estimation of scalar
values in one form of "structure extraction" and the scene transformation opera-
tion of slice interpolation. (45) and (41) discuss the different 3D medical imaging
rendering processes.

All 3D medical imaging techniques assume scene space consists of a 3D array


of values (27). The 3D array is actually stacked slices of scanner generated data (see
figure 2.1) with inter- and intra-slice thicknesses. This format will remain following
an 3, scene transformations.
Two processes that must be accomplished to perform structure extraction are
segmentation and boundary detection. Segmentation partitions object space into
objects of interest (which may or may not be surfaces) and the remaining volume.
Boundary detection or surface tracking are techniques that locate surface boundaries
(e.g., organs or bone) during image segmentation (40). These processes are referred
to in the discussion of surface methods, but are mentioned here because they are
done in both volume and surface methods.

2-5
The remainder of this section provides a very brief overview of the contour and
volume methods and discusses the surface methods in more detail. For a complete
discussion of these different approaches see (19) and (41).

2.3.1 Contour Methods The earliest approach to 3D visualization creates 1D


contours (1D units). contours represent the border of the object of interest. Herman
et al. (26) gives an excellent description of contours, along with a figure that clearly
depicts a common ID unit-based rendered surface. A contour corresponds to the
surface border of the object on one particular 2D slice of the data. A contour is
usually drawn as an approximated curve. When the contour lines for all the slices
are rendered, the appearance of a 3D object surface is created. Contour lines are
extracted manually, automatically, interactively, or by a combination of the three
methods.
The obvious disadvantage of contours is the poor representation of the surface,
because the only information depicted is in the plane of the slice and within the
plane, only the surface boundary of the object. However, 1D unit-based images
are generated and rotated rapidly. Due to the latter fact, they have found wide
acceptance and use in the medical imaging community.

2.3.2 Volume Methods These techniques are volume oriented because they
use 3D volume units as display primitives. Volume techniques differ from surface
and contour methods by the amount of data that must be stored and processed
during image computation. Additionally, volume methods typically preserve data
continuity between surface boundaries, whereas surface and contour methods do
not.
Surface rendering algorithms assume data consists of objects with thin surfaces
in a, volume of air (i.e., surfaces are easy to find, with little noise), whereas in reality
most objects have fuzzy (thick) borders and there is much more than just thin
air between surface boundaries. Volume methods preserve fuzzy borders and inter
surface material by avoiding simple classification schemes that assign binary values
to voxels indicating the voxel is in or out of the image to be rendered (33). The
phenomenon that a voxel may contain more than one scalar value is termed a partial
volume artifact. Volume methods allow percentages of different scalar values as
well as colorattenuated light and /or transparency to be assigned to a single voxel

2-6
(16). Each voxel contributes to the final image based on these percentages, thus
reducing the effect of partial volume artifacts. The final color of a pixel becomes the
contribution of all voxel values lying along a ray's line of sight or along a projection
path. Colors are weighted by transparencies and attenuation. Since every voxel
contributes to the final image in this way, volume methods capture transitional areas
between surface boundaries that might otherwise be missed by surface methods.

2.3.3 Suiface Methods Surface methods attempt to reduce the volume of data
to surface boundaries by depicting these boundaries by common graphics primitives
such as polygons, patches, or points (2D units), Surface methods process a small
number of slices at a time, hence they had dominated imaging algorithms for many
years since computing power was not sufficient enough, until recently, to handle the
entire volume of data at once. These techniques consist of tiling, surface tracking
methods, and cell interpolation approaches.

2-7
Tiling
Tiling (tessellation) methods take as input contoc.s created from any of the
contour approaches. The next step usually filters the contour data by smoothing or
re-sampling and then polyhedra such as triangles or quadrilaterals connect adjacent
contours. Smoothing is done to better approximate the curve nature of contours.
One heuristic-based method (42) uses B-splines to approximate a closer fitting con-
tour, so contour smoothing is unnecessary. The following paragraph descri" s some
of the tiling techniques developed in the past.
Tiling methods fall into two general classes, optimal based and heuristic based.
The optimal based solutions (23) and (30) apply graph theoretic methods to derive
what the authors consider the optimal triangulation between two adjacent planar
contours. The major disadvantage of the optimal tiling methods is the long search
time required to find the best triangulation. However, the process is entirely auto-
matic. Since speed is an issue in medical imaging, many other methods were devel-
oped based on hueristics to achieve a faster tessellation, adding interactive assistance
if needed for ambiguous cases (5),(42), and (24).
One major advantage of tiling is that it produces conventional graphics geo-
metric primitives that can be rendered by applying standard reflection and shading
techniques. In addition. as with contour methods, rapid viewpoint, changes are pos-
sible and the data size in the final imaged data set can be quite small compared to
the original volume of data.
Gross inaccuracies in an image can occur if contours are not well-formed with
respect to each other. For example, this can happen when more than one contour
is formed on a scan plane to represent a surface. The most well-known remedy
is interactive editing, although it. is time consuming and still error-prone. Even
without interactive editing, automatic edge tracking to find the contours can be
too slow for most applications. Speed of contour formation is proportional to the
number of structures in the data set. Another major disadvantage is that tiling
based on contours results in loss of essential information because contours do not
contain enough gradient data to represent. the actual surface.

2-8
Surface Tracking
Surface tracking methods generate a surface as a set of cuberille faces. Recall
from chapter one that a cuberille is a dissection of 3D space into equal size cubes by
three orthogonal sets of equally spaced parallel planes. This is a natural t.:tension
of the 2D space dissection forming quadrilles, or "square-shaped pixels" (4:34). The
output primitives are planar polygons formed by connected cube faces approximating
a surface of interest.
The first step usually accomplished in surface tracking methods is to modify
the volume to form cubes. This is done by interpolating values in the components
needed (19:327)., For example, Artzy et al., (1:19), Artzy (2:6), Udupa (44:220-221)
linearly interpolated in one dimension so their input data would have an interslice
distance equal to the resolution of the original 2D slices. This works as long as the
resolution of the original 2D slices is square. If not, interpolation in two or three
dimensions might be necessary.
Prior to rendering the cube faces, two tasks must be accomplished. First, the
voxels must be segmented into those being in the object(s) of interest (1-voxels) or
out (0-voxels). This binary classification forms a 3D binary volume (45:49). Next,
the surface boundary(s) located between 1- and 0-voxels must be located and display
elements connected. The term surface tracking is derived from this process of locating
the surface boundary from a binary volume.

Artzy et, al. (1) developed a surface tracking algorithm that reduces the chal-
lenge of finding connected voxels representing the surface boundary to a graph traver-
sal challenge. Voxels are first segmented using binary classification. Next, boundary
detection is accomplished. Nodes of a directed graph, C, then correspond to voxel
faces separating the object under interest, from all else in the scene. The authors
prove that connected subgraphs of G correspond directly to surfaces of connected
components of the object. TO find the surface boundary, a subgraph of a digraph is
traversed.

2-9
Cell Interpolation Approaches
Cell interpolation methods generaLe polygonal elements by analyzing compu-
tational cells. Recall from chapter one that a fom utational cell is a parallelepiped
such that four cell vertices are voxels in one slice and the other four are voxels in an
adjacent slice - see figure 2.4, The major difference !-etween these methods and the
cuberille methods is that the cell interpolation methods analyze how the data varies
between voxels to determine where the surface lies versus assuming only constant or
linear variation.

F e . uicomputational

'Voxels formling cell vertices

Figure 2.4. C omputational Cell

There are currently four types of cell interpolation approaches. These algo-
rithms are the marching cubes method developed by Lorenson and Cline (34). the soft
objects method by Wvvll and McPheeters (52), the "gradient-consistency heuris-
tics" by Wilhelms and Gelder (50). and the cell subdivision techniques also described
by Wilhelns and Gelder (50).
Each method follows a two step process. First, the volume of data is segmented
by classifying each voxel as eil her I or 0 (or in the case of the soft objects method as

2-10
hot or cold). A voxel is . f its scalar value is greater than an iso-value (threshold),
else it is assigned a 0. The second step is to determine if a cell has both 1 and 0
vertices and if so generate polygons within the cell to approximate the portion of the
iso-surface that passes through the cell. An iso-surface is formed by connecting all the
polygonal elements to form a 3D mesh, such that the surface inter,-cts approximately
the same (iso) scalar value or range throughout the sample data. Polygon vertices
(not to be confused with cell vertices) are determined by linearly interpolating 3D
coordinates between the 1- and 0-voxels of a cell. The coordinates are interpolated
to the iso-value. The four methods differ by how the polygons are formed within the
cells.
The term ambiguous cell must be defined before discussing the four cell in-
terpolation methods. A cell is termed ambiguous if more than one topology can be
chosen for it. A topology is the polygon formation within a cell. Durst (18) noticed
holes can result from the marching cubes algorithm described by Lorenson and Cline.
Holes can be caused by improperly forming polygons between two ambiguous cell
faces.
The term ambiguous cell was defined by Wilhelms and Gelder (50). An am-
biguous cell face is defined as "a cell face that contains a diagonally opposite pair
of positive vertices [1-voxels] and a diagonally opposite pair of negative vertices [0-
voxels]" (see figure 2.5, obtained from (50)). By looking at figures 2.6, 2.7, and
2.8. the reader can see that the a,,biguous cases are 3, 6, 9, 12, 13, and 14.

2-11
+

I S

- 4
-

Figure 2.5. Ambiguous face - intersection point connectio, choices

Marching Cubes

The marching cubes Li :rithm was developed to alleviate the necd of using
cubic voxels, i.e., 3D data " with reduced resolution in one di,,ension" (7:345).
Binary classification of cube vertices creates a total of 28 = 256 possible cell vertex
classifications. Py analY . the geometry of the different cases, the total number of
unique cases can be red:,ced to only 15 (see figures 2.6, 2.7, and 2.8). The other
241 cases are reduced to the 15 by symmetry and appropriate rotations. Wilhelms
and Gelder (50) call this approach the major case table looL-up iiiethod because a 256
element table must be pre-set to indicate the transformation of cases to the unique
15. The signs at. the cell's eight vertices are then used as an index into the major case
table. Once the appropriate classified cube case is determined, another pre-set table
ent, indicates the triangle formation. The triangle formation is somewhat arbitrary
since intersection points (points approximating where the iso-surface intersects a cell
edge) can be connected in many different ways for most of the 15 cases. Only certain
connections make sense with most of the cases. However. there are six cases that can

2-12
cause serious inaccuracies if improperly connected (which is discussed in the next
section).

2-13
6 7

2 -73

1 Case I
Case 0

case 2

case 3 case 4

Figure 2.6. First 5 cases of 15 cube vertex classifications and triangulations


I I

case 6
case 5

case 7

case 8 case 9

Figure 2.7. 5 more cases of 15 cube vertex classifications and triangulations

2-15
case 10 case I1I

case 12

case 13 ca-se 14

Figure 2.S. Final 5 cases of 15 cube vertex classifications and triangulationis

2-IC
c

C C

CC

case 6 case 6
Front facial average
Front facial average greater than threshold
less than threshold, so connect hot vertices
don't connect hot verices

Figure 2.9. Facial averaging example

Soft Objects

Wyvill and McPheeter's soft objects method is almost identical to the marching
cubes algorithm., The actual 3D scalar field is fabricated rather than obtained from
scanners, but the data representation is still a 3D regular data set. Key points are
specified to reduce the data set. Additional points are estimated as needed by a
cubic function that uses a radius of influence to determine the key points needed in
the estimation function. The most significant difference between this method and
marching cubes is that this method uses a dynamic. simple technique to polygonize
an ambiguous cell.

The soft oojects method forms polygon vertices on an ambiguous cell face by
analyzing the four cell face vertices. The method assumes the value at the center of
the face is approximated by averaging the four vertex values. Then, if the averaged
value is greater than the iso-value, the positive vertices are connected. For example,
see figure 2.9. L, his way, two cells sharing ambiguous faces are always consistently
polygonized - thus no holes result.

2-17
This facial averaging method provides only a rough estimate of center face
values. A different form of estimation may be required to obtain a more accurate
center value estimate. Even if the estimate is not veLy accurate, Wilhelms and Gelder
claim the facial averaging method guarantees continuity, i.e., no holes appear in the
image. This assertion is based on their facial plane principle :

If the method of disambiguation for ambiguous faces employs only values


in the plane of the face, and is invariant under rotations and mirror
images, then the isosurface as defined by topological polygons will be
continuous.

They refer to the (possibly nonplanar) polygon generated within a cell as a topo-
logical polygon because it specifies "the topology of the isosurface within the cell."
This assertion makes intuitive sense because the same facial average is calculated for
shared cell faces. This results in shared vertices between cells for that face.

2-18
Gradient Consistency Hueristics

The gradient consistency hueristics were developed by Wilhelms and Gelder to


compute a better estimate of the center face value, assuming that simple averaging
does not estimate the center value accurately enough (50). The authors use these
methods to handle ambiguous cases in the same manner as facial averaging. The
authors use the gradients at the four face corners to obtain the center estimate.
Gradients indicate how rapidly the iso-surface is changing at a particular point.
The magnitude of the gradient is used as an approximation to the normal in the
shading calculations, therefore it is already computed at each voxel. The two gradient
consistency hueristics developed by the authors are the "center-pointing gradient"
and "quadratic fit."
The center-pointing gradient is the gradient component of a cell vertex directed
towards the cell face center. This gradient is obtained as the dot product of the vertex
gradient with a unit vector in the direction from the vertex towards the -enter of the
face. T-No univariate quadratic functions are determined by the least squares error
method to estimate the center scalar value along both diagonals of the face. These
two estimates are then averaged to yield a final estimate. This estimate is then used
to determine the cell topology, as in the soft objects facial averaging method.

Two assumptions are made in this method. First. the center-pointing gradients
are assumed to approximate the deriivatives at each corner. Second, the quadratic
functions are assumed to exactly fit the two endpoints. The gradient assumption is
typically used for determining the normals for shading as well. The second assump-
tion is made in many estimation methods to insure the estimation process returns
the correct value at known points.
This method only analyzes the face corner sample values, similar to the soft,
objects method. It. should choose the topology better than the soft objects method
only if the tuderlying scalar field function is quadratic along both face diagonals.,
The quadratic fit method is similar to the center-pointing gradient method,
except here the center face value is estimated by a single bi-variate quadratic func-
tion. The least squares error fit uses all four face corner sample values to estimate
parameters of the function. This method chooses a more correct topology than the
center-poitaing gradient if the underlying scalar field function is quadratic across the
entire face.

2-19
The gradient consistency heuristics make minor improvements to the soft ob-
jects method of resolving ambiguous cells. A disadvantage of the heuristics methods
is that they assume the scalar field function is locally quadratic. If wrong, this as-
sumption can generate inaccurate topologies within cells. The only way to knowv if
the topology is inaccurate is to know the local variation of the scalar value function.
However, obtaining this knowledge might be too computationally costly to be of any
benefit.

2-20
Cell Subdivision Techniques

Cell subdivision techniques are implemented for two primary reasons. The first
is to increase image fidelity. The second is to resolve ambiguity of cells. Image fidelity
can be increased because the data resolution is increased. Cell subdivision is the same
method employed by Cline and Lorensen's (8) dividing cubes algorithm. In dividing
cubes, cubes (cells) are divided until the resolution of point primitives is reached.
In cell subdivision, instead of subdividing cells to create point primitives, cells are
subdivided to create sub-cells. These sub-cells are then treated as the previous
original cells were, i.e., the iso-surface is represented within them by triangles. These
sub-cells are much smaller than the original ones, hence data resolution has increased.
Because of this increased data resolution, the surface can be approximated closer to
the actual surface by the smaller triangles. However, the quality of the final image
depends on how the scalar values are estimated at the sub-cell vertices. Therefore,
the significant challenge with any subdivision technique is how to estimate values at
the sub-cell vertices.
According to Wilhelms and Gelder, another way to resolve ambiguity in cell
cases is to create sub-cells within original cells and polygonize the sub-cells as in the
soft objects method. New cell vertex scalar values are estimated by a re-sampling
fumiction. In \'Villhelms and Gelder's implementation, they subdivide each computa-
tional cell and apply an estimation function to determine new scalar values at. the
points derived from the subdivision. The two estimation functions they use a.re tri-
linear and tricubic. The major case table lookup method (marching cubes) is used
to process the sub-cells, and ambiguous cases are handled as in the facial averaging
technique. Tricubic is more coinputationally intensive than trilinear. but it produces
more accurate iages for their artificially created volumes than any of the other cell
interpolation methods. Image quality improves because the tricubic method consid-
es sampIle (lat a in a large neighborhood (the surrounding 6,4 voxel values) and does
not assume linear variation. On the other hand, trilinear only analyzes the eight cell
vertex values and assunes a linear variation along each of the three axes.

2-21
Note that all nonempty1 cells must be subdivided to ensure continuity between
faces. If not, one large undivided nonempty cell face may share a face with several
sub-cell faces.
Besides just resolving ambiguous cases, cell subdivision is also a good method
to use if a smoother looking image is desired. However, depending on the subdivision
size, the resulting number of polygons may be very large. A typical set of from 50 to
100 brain MRI or CT slices can result in over 500,000 polygons for certain iso-values.
Subdividing by a factor of just two could increase this to over two million polygons.
In this case it might be more beneficial to use Cline and Lorense,'s Dividing Cubes
algorithm (8).

'Nonempty means that a cl.sification of the vertices derive- unique cell case 0 in figure 2.6.
That is. a nonemnpty cell is one such that the surface is determined to not pas.' through it

"2-"2"
Surface Methods Conclusion
Both the cuberille-based methods and the cell interpolation methods produce
images that appear to represent the surface of interest. However, the most important
issue is not appearance but rather the accuracy of the methods. An argument could
be made that the pre-processing step of creating cuberilles produces more accurate
results because of the increased resolution. However, new slices are normally created
by simple linear interpolation, whereas a higher order interpolation might be more
accurate. Also, once the cuberilles are formed, constant variation within cells is
assumed during surface formation.

Cell interpolation techniques using subdivision and estimation functions also


estimate new scalar values within the original computational cells and can easily
go beyond input pixel resolution by increasing the subdivision factor. In addition,
according to Upson and Keeler (46), the variability of the scalar values within cells
..is a more accurate representation of the real world and allows for smooth, con-
tinuous representations of even small datasets." Cell interpolation methods assume
variability within cells to create new sub-cells. They also assume scalar value varia-
tion within the new sub-cells (although this latter variation is always linear - recall
the process of finding surface intersection points along cell edges).
An advantage shared by the cuberille-based and the cell interpolation methods
is that tile create a display list of polygons. This display list can be saved for fast

re-display at different viewl)oints using traditional lighting and shading models.


A major disadvantage to the marching cubes cell interpolation algorithm is that.
holes can appear in the rendered surface (18). Yet, with disambiguation methods
employed, the ambiguous cases are corrected to a certain degree. Cell subdivision
methods can also improve image quality as well as the accuracy of surface represen-
tation. Both image quality and accuracy of surface representation depend directly
on the accuracy of the estination fiction.

2.4 Chlptf r Stulfiary

3D medical imaging topics relating to this thesis were presented. Specifically.


this chapter first discussed tie data representations used in 3D medical imaging.
Regular data grids are assumed by all 3D medical imaging methods because the
output from -nedical imaging modalities is regular. The two )rimary data models

2-23
include voxels and cuberilles. Next, the common 3D medical imaging transforma-
tions were summarized. The remainder of the chapter described the 3D medical
imaging methods - contour, volume and surface. Surface methods were covered in
great detail because surface methods were iiplemented to accomplish the purpose
of this thesis. The first primary surface method discussed is surface tracking. Since
these techniques use the cuberille data model, surface tracking methods usually es-
timate new slices between existing slices to form cuberilles. Although a surface
tracking method was not implemented iii this research, the task of slice interpola-
tion was accomplished. Finilly, cell interpolation surface methods were discussed.
These include the marching cubes algorithm, the soft objects method, the gradient
consistency heuristics, and the cell subdivision techniques.
The focus of the next chapter is to explain how an estimation function can be
derived based on the statistics of the underlying scalar field data.

221
III. Kriging Theory

Kriging is a geostatistical estimation method name6 after a South African min-


ing engineer, D. G. Krige. The theory was developed for the purpose of estimating
ore reserves in mining. However, kriging can be used to estimate samples in any situ-
ation where estimates depend on nearby known samples based on spatial or temporal
position. Kriging is significant because it is the optimal linear estimator.
This chapter presents first the background and purpose of kriging, followed
by the derivation of the ordinary point kriging equations. After these equations are
derived, different kriging categories and types are summarized. Lastly, an overview
of structural analysis and isotropy is presented. The order of topics was chosen to
present the most common form of kriging first, and then building on this foundation,
to introduce other kriging methods.

3.1 Background
Krige developed the basic theory, but ,. French engineer named Georges Math-
eron and his colleagues developed the rigorous mathematical theory of kriging (35:602)
and (9:625). Prior to kriging. estimation methods used in geo-statistics made several
simplifying and usually Invalid assumptions. The most erroneous assumption was
that variances between data samples is constant. This assumption made the other
estimation methods very error-prone. Krige pointed out that to get a more accurate
estimate, variances between the prospective blocks and the core samples must be
taken into account. Kriging is primarily used in situations where there is expected
to be some dependence between data measurements at different locations. Its use in
discovei ing deposits in various mining pursilts is well known and documented; see
(13:70-71).,
Kriging is a process that derives a geo-st.atistic. Geo-statistics differ from
classical statistics in the variables tied. Recall that a statistic is a function of random
variables. A geo-statistic is a func- ion of regionalized variables. Random variables
model chaotic processes. Regionalized variables model spatially dependent. natural
phenomena. According to Matheron (35). regionalized variables are characterized
by three qualities. The first is localization. Regionalized variables are localized
within a Support. A support is the volume of a sam)ple. consisting of geometric

3-1
size, shape, and orientation. In the geo-sciences, an example support is a drill core.
The second quality of regionalized variables is that they may exhibit continuity
within the region of a support. Stadsticai continuity means the sample values do
not deviate significantly from each other. Thus, they are not random in nature,
but show some kind of order. The third characteristic of regionalized variables is
anisotropies, discussed later in this chapter.
Kriging is a modified form of a multiple linear regression model with parame-
ters estimated by a technique similar to the method of least squares (37) and (15).
Kriging uses weighting functions based on distance to coml)ute the desired data
value. The method operates on the assumption that data points closer to the target
should be weighted heavier in the estimation calculation than those further from the
new point. For example, in figure 3.1 points 1,2, and 7 would be expected to have
more influence on the estimate than points 5 and 6. This weighting strategy has

05
03 3

0
2
0 t<( Estimate

Figure 3.1. Kriging Example

provet to be very accurate in the field of geo-statistics. Davis sums up the goal of
krigiug in t.he following sentences

3-2
There are an infinity of other possible combinations of weights that could
be chosen, each of which will give a different estimate and a different
estimation error. There is. however, only one combination that will give
a minitnum estimation error. It is this unique combination of weights
that kriging attempts to find.

The goal of kriging is to estimate an unknown value at a particular place


within a known neighborhood of points using a linear combination of computed
weights and known sample values. Optimal weights are determined by solving a
system of linear equations produced by two conditions placed on the kriging linear
combination. These constraints make the value derived from kriging the best linear
unbiased estimator (B.L.U.E.) (13). The constraints are:

* unbiasedness by setting the expected estimation error to zero i.e., removing


any "systematic error"
" minimum estimation error variance.

These conditions insure optimality and insure that "no other linear combina-
tion of the observations can yield estimates that have a smaller scatter around their
true values" (14:385).
Why use a linear estimator versus a non-linear? According to Delfiner and
Delhoimme. "The major difficulty with non-linear estimators is that they involve
parameters or characteristics that cannot be inferred from the data" (15). Even if
the sampling function is locally non-linear, this is taken into account by the drift in
universal kriging. which is discussed later.
The basic purpose of kriging for this research is the same as that for the
estimation fia'ctions explored iw Williclms and Gelder (50) - given a flew point
within a neighborhood of known sample points and associated values, estimate a
value at the given point using a combination of known sample point values. The
kriging equation esl imates a value as a distance-weighted linear sum of known sample
points. The termn linear refers to just the linear suni. it does not indicate that the
data varies linearly. Kriging can nodel all forms of highet order data trends such
as quadratic and cubic. This capability is discussed in later sections. The next few
sections explore ho" kriging calculates data weights and show that the conditions
assune(l about the(data make kriging the optimal interpolator for data sets derived
from natiral phenomena.

3.3
3.2 Krigzng and Least Squares

Kriging primarily differs from least squares in tie type of variables used in the
linear equation. Tile variables used in least squares are assumed to be independent
random variables, whereas those in kriging are regionalized variables. Recall from
a previous definition that regionalized variables model natural phenomena. These
types of variables assume the data is localized and exhibits continuity, whereas phe-
nomena modeled by random variables exhibit chaotic behavior. No simple methods
like least squares can be applied to such variables because no tractable deterministic
functions can be found that describe the complex variations in regionalized variables
(14). The next section discusses how the dependence of regionalized variables is
captured in the kriging linear regression model in tile simplest kriging case.

3.3 Derivation of the Ordinary, Point Kriging Equations

My discussion of kriging begins with ordinary point kriging because the other
forms are modifications of this method (ordinary and point are defined in the next
section). Tile discussion begins by stating the goal of kriging. followed by an explana-
tion of the constraints on the kriging equation that produce a system of equations.
The system of equations are solved to yield the weights 'in the kriging equation.
The equations presented below were obtained from a number of different sources.
primarily (12). (14). (17). (9). (15). (13). (48). (11). (31), and (6).
In the following equation the goal is to estimate Z. the unknown value at the
known position p in the neighborhood of known points p, and known values Z,(p,).

The Z,s are the regionalized variables with the parameter being an n-dinensional
point and the r,*s the weights. The weights are chosen to satisfy the following two
conditions that make 7. the B...

* E( - Z) = 121
o E(Z - Z)' minimumlir [31

where 7 is tihe value being estimated at p and Z is the actual value at point p. The
estimation error. 7, - Z, is a measure of the dissimilarity between the two variables
7 and Z. E(;7 - Z)2 is Ihe nean square error and h, is the expected value or mean.

:1 I
Using the following equality from the definition of the variance, V, (37:89),
E(Z - Z)2 = V(Z - Z) + [E(2- Z)12
[3] can be re-written as
( 11(2 - Z) minimum [4]
where a' is the estimation or error variance. This is important because it points out
that even though condition [3] states minimum mean square error, it is equivalent
to minimum estimation error variance.
Next. the system of kriging equations are derived from conditions 121 and [4].
This system is similar to the set of simultaneous linear equations (normal equations)
produced in linear regression that are formed by setting the partial derivatives of the
unknown parameters to zero (21). Before the equations can be derived, the above
two conditions are expanded and changed into more quantifiable constraints.
Modifying condition [2] above is straightforward
E(Z - Z) = 0
E(2) - E(Z) = 0
Then recalling [1], substitute into the above and get as an additional constraint
EjZwjZ(p,) - E(Z) = 0

>Zv,.,(pz) - n"(p) = 0

E = [.51

where nt() is the mean or first moment. Unbiasedness in the estimate is assured by
insuring that the weights sum to 1. (13"238)
The eslimalion error variance is ((6). (13). and (9))
- = v,',(),.p) - K (p,) - E .,,K,(p,.p,)[ '

where K(m.n) is the covariance between point m and poiiit n and 1) is the n-
dimensional point where the estimate is computed. Jo,,rnel derives similar esti-
mation e,-ror variances (29), but this equation is considered the general unbiased
linear estlimator derivable by expanding the variance of the linear combination of
regionalized variables in equation [4].
The covarianee between the variables is modelled by a function called the seni-
variogram. In most cases the semivariogram is ntiknown and must be determined

3-5
by a process called structural analysis. For the present. assume the semivarogram is
known and is represented as ((m,u) = mn. The semivariogram represents the av-
erage difference squared between the values at points m and n. The main parameter
used in semivariograms is the distance between points in and n. The semivari-
ogram models the dependency of data values based on how far apart they are from
each other. The spatial distribution of regionalized variables is accounted for in
the semivariogram. The semivariogram gives the correlation between san-ple values
a geometric meaning rather than a probabilistic meaning. Structural analysis and
the semivariogram are discussed in more detail in a later section. As a result of
substituting the semivariogram in place of the covariance. [6] becomes
C2
( */Z~2 ~ ip
-ZLCWh 7

Now that condition [1] is more quantifiable in terms of [7]. it must be minimized
to satisfy the minimum estimation variance constraint. Since there is a constraint
[5], n unknowns and n + 1 equations would result from minimizing this system.
Therefore a Lagrangian Multiplier is added to equalize the system. Minimization
is done by taking the partial derivatives with respect to the weights an( the La-
grangian Multiplier tj and setting the resulting equations equal to zero. 1 his yields
the following complete ordinary point kriging system:
It

Ec,' + 7=
I (ii= 1. 7.) (8
f
J=i

The Methodology chapter will show these equations in expanded matrix form.

Tile systemn presented above is ordinary point krigiig. Due to 'he consi raints [21
aiidl [.1]. this system will determine optimial weights to subsitule back into eq)aatio
[1]. Z is the B.L.U.E. because of the optimal weights. i.e.. no better linear estimate
can be derived. This system is only one of several types of kriging possible. Other
types and their differences from this system are discussed next.

3.4 Krqiiny CalyilnOirs and 7htips

The kriging literature describes two broad categories of kriging and three tees
of kriging. Thr categories differ ba.sed on estimation region. The t-pes differ based

3-6i
on assumptions about the sample data. Either of the two categories can be used in
any kriging type.

3.4.1 Kriging Categories The two categories of kriging are roint and block.
Point kriging was discussed in the preceding section. It is employ, When the goal
is to estimate a value at one particular point.
Block kriging estimates a value for a region instead of at a single point. There
are two block kriging methods. The first uses point kriging repetitively to estimate
several values within the block and then averages the results to get one value. The
second method derives a new set oi equations using a modified covariance function
of the K() terms in equation [6]. The second method involves computing a double
integral that, evaluates the area of the block in question. The problem with the
second approach is finding an explicit analytic form for the integral; hence, the first
method is most often used for block kriging (1 )1.

3.4.2 Kriging Types There are three primary types of kriging discussed in
the literature - simple, ordinary and universal. These techniques differ in their
assumptions about the behavior of the expected values or means of the regionalized
variables E(Zj) (15). see figure 3.2.. Each of the three methods can be developed so
the final kriging system estimates points or blocks.

'Some authors conib)in poii and block kriging into one form in which the area integral reduces
to a point in the cast of pomt kriging (9:626)

3-7
Krjging Means of Regionalized Means of Regionalized
lype Variables are Known Variables are Constant

Simple Yes Not applicable

Ordinary No Yes

Universal No No

Figure 3.2. Table of Kriging Types (All three can estimate points or blocks)

Simple Kriging
This type of kriging, as the name suggests, is the simplest form of kriging
- even simpler than the above ordinary kriging system. However, this nethod is
seldom employed. The method is termed "simple" because sample means at known
locations ,tre assumed to be known prior to kriging. The means are stated as
E(Zj) = mi,i = 1,...,n and E(2) = in
where mi is the mean of the i'th regionalized variable and m. is the mean of the
estimator regionalized variable.
This assumption modifies the kriging equations derived above because the un-
bias constraint changes. The ordinary equations derived above left out one variable
important to the simple kriging equations. This variable is called a shift parameter.
A (29). In ordinary kriging, A = 0. The shift parameter modifies the estimation as
n
Z=A + Zi (p,
117,

Now the unbias condition becomes


E(Z - Z) = E(Z) - E(Z) = E(YZuw,Z,,(p) + A)- nz
Journel (29) shows that this unbias constraint changes the estimation equation to

Z =n+ Zwi(Z, - ?n')


i=1
This equation indicates that the unbiased estimator 2 is determined as a linear
combination of the residuals, (Z, - ni,).

3-8
This assumption also simplifies the minimum error variance condition [4]. In
essence, this form of kriging reduces to classic linear regression (29) and (12). Simple
.:riging is seldom used because the means of the regionalized variables are usually
unknown.

Ordinary Kriging
Unlike simple kriging, ordinary kriging assumes the mean of each regionalized
variable is unknown. Also unlike simple kriging, ordinary kriging assumes each mean
is the same. A constant mean is more commonly referred to as a stationary mean
(9:626). The ordinary system of equations were derived above to estimate points.

3-9
Universal Kriging or Kriging with a Trend

Universal kriging also assumes unknown regionalized variable means; however,


it differs from ordinary kriging because it assumes the unknown sample means are
possibly different, or non-stationary. Sample means are also referred to as first-
order moments. Stationarity assumptions for the second-order moments can also
impact the type of kriging used (see (9) and (11)), but will not be considered ,in
this effort. The term non-stationary also refers to drift or trend, indicating the
means of the regionalized variables differ from sample point to sample point within
a surrounding neighborhood of other sample points. There are two plrimary methods
of kriging in the presence of local drift - universal kriging and the method of intrinsic
random functions. According to Cressie (9), universal kriging, simply extended from
ordinary kriging, is most often used because the method of intrinsic random functions
is impractical to apply.

When using universal kriging, the first process becomes estimation of the re-
gionalized variable means at the sample points using a local neighborhood of known
sample values to determine if the means are constant from sample point to sample
point. If data has global (iegional) drift, there appears to be a definite pattern or drift
in the sample values over a larger area, usually much larger than the neighborhood
size used in the kriging system. If drift, is local, it occurs within the neighborhood.
Figure 3.3, adapted from (14) and (13). shows regional drift as a line placed through
the data poinls.

If global drift exists. the regionalized variables are now viewed as being com-
posed of two parts - the drift and the residual. The drift of a regionalized variable is
its expected value (mean) at a point p, within a certain neighborhood. This experi-
mental or computedi meat is called drift if it varies from point to point. The residual
is calculated by subracting the drift from the actual measurenieqIt. For example.
assume an experimental drift is calculated for every known point. call it ?n(Z(p,)).
Then the residuals are calculated as

Y(P,) = Z(p,) - ,,,(Z(,,)).

2David shows that the experiniital drift. can be calculated by estimating the drift coefficients as
a linear combination of the a\,ailable data. This is another multiple linear regression that tq pically
assumes simple unbiased estiiation derived from a least-square inethod (13) The drift equation
is shown in the ntxt. paragraph.

3-10
Z(X)

Global Trend and Local Stationary Mean

Z(x)

Local Trend and Global Stationary Mean

Figure 3.3, Global and Local Drift

3-11
There are three primary steps in estimating points or blocks in the presence of a
global drift. First, the drifts are estimated at each point and residuals are computed.
Then, the residuals are used as stationary regionalized variables in a simple, ordinary
or universal kriging system (depending on the local means). Lastly, the estimates
derived by kriging the residuals are added back to the drifts to get the final estimate.

If local drift exists, then universal kriging is used (13:267). The ordinary point
kriging equations developed in the last section must be modified in the presence of
local drifts to yield the system of universal kriging equations These modifications
are described below.

David (13) expresses the drift. at point p as


k
rm(p) = E(Z(p)) = Zdtf'(p) [9]
1=0
Equation [9] is normally represented as a finite order polynomial in universal kriging.
The f'(p)*s are k + 1 known functions, usually monomials, and the d,'s are the k + 1
unknown drift coefficients.

The kriging system of equations change in the presence of drift because the un-
bias constraint changes. In ordinary kriging the unbias constraint forces the weights
to sum to I because the mean ni(u) is constant, producing equation [5]. Now that
rn(p) is no longer constant but takes into account drift from equation [9]. the unbias
constraint results in

E(2 - Z) = 0
E(ZZ(pi)) - Zdif(p) = 0
a k

-wi.f'(p 2 )_ f=(p) ( = 0.1 ..... /,) [10]

Notice the drift coefficients di have dropped out of the constraint. Thus the
universal system is independent of the drift coefficients, but still insures unbiasedness.
Since this condition insures unbiasedness regardless of the unknown drift coefficients
di, the term universal is used to denote the system of equations that result.

This constraint (equation [10]) adds k+ 1 more equations to the minimum vari-
ance condition, thus k+1 additional Lagrange Multipliers (711) are needed (9).(15).and
(17). After the partial derivatives of the equations are taken with respect to the n
weights and the k + I Lagrangian Multipliers and set to zero. the resulting universal
kriging system is

3-12
n k

EUr,, + Z-ii.f'(P,) = iv (i = 1,...,n) minimum estimation


3=l 1=0
variance [1
n

Elwf'(p,) = f'(p) (I = 0, 1, .. ,k) non-bias [12]


t=1

There are several unknowns in this system of equations that must be esti-
mated. The unknowns in the universal system are the order k of the polynomial
f.(p), the drift coefficients di, and the size of the neighborhood used to determine
the drift. These unknowns are determined during structural analysis. The drift co-
efficients dl can be found along with the weights in the kriging system (14:394) or
by a least-squares method (13:272). The order k is usually 1 or 2. If the means of
the regionalized variables are the same, k = 0 and equation [10] reduces to equation
[5], which is ordinary kriging. If the order is 1 this means the 0th order term or
the constant will be included as well as the first order terms. In two dimensions
this is the x and y terms. If k 'is 2, the 0th, 1st, and 2nd order terms are included
in the drift. The first order (k=l) polynomial associated with linear drift in the
neighborhood is : m(p) = do + d1 Xi + dX 2i
and the second order polynomial associated with quadratic drift is

do + dlX, + d2 X 2, + daX1, + d4 XiX2i + dsX 2 ,


?l(p) =

where X1, and X 2, are the first and second coordinates of the i'th known 2D point.
in the neighborhood (14:394). As stated earlier in this chapter, any order drift can
be modelled by kriging - simply modify k in equation [12]. If a polynomial drift is
not observed during structural analysis, other types of drifts can be easily modelled
as any type of function of the geometric coordinates.

-3.5 Shuclural A wiiysis

Structural analysis is "the process of attempting to simultaneously find satis-


factory representations of the semnivariogram and drift expression." (14:245). This
process also determines the optimal neighborhood for the estimation. The neigh-
borhood is the sample values in the kriging linear sum that are all within a certain
distance of the estimated value. I term neighborhood size the nunber of sample
values in the kriging sum.

3
1)aVis (14) and David (13) present tie universal kriging equations in expanded matrix form.

3- 13
The semivariograrn, drift, and neighborhood all influence each other and char-
acterize the notion of localized continuity within a sample volume. The goal of
the process is to find a model semivariogram that models the spatial correlation of
sample values within a local zone of influence (neighborhood).
Structural analysis is usually performed prior to kriging. Prior to understand-
ing the process of structural analysis, the semivariogram must be defined.

Semivariogram Definition
The semivariogram is a graph and/or formula (14) that

...represents the spatial variability of data ... can be thought of as an


average difference squared between data a given distance apart in a given
direction... [and] provides a quantitative value for the range of influence
of a sample in any direction (31).

The semivariogram is used in the kriging system of equations as an approx-


imation to the co-variance between sample values. The co-variance measures the
inter-dependence or correlation of random variable occurrences, whereas the semi-
variograin measures the spat ial dependence of regionalized variable occurrences based
on the distance from each ot her.
There are two major types of semivariograms used in kriging - experimental
and model. The experimental semivariogram is used to estimate the variance of
differences in the sample data. The experimental semivariogram is an estimator of
the model senivariogram. Once aii experimental semivariogramn is computed. it is
compared to known model semivariograms to select the closest match. The reason
for using a model semivariogram is discussed in the section Models. The remainder
of this sect ion will describe how the experimental semivariogram is derived.
The semivariograni ^,,(h) is often improperly called the variogram. It is termed
the semivariogram beca'ise it is half the variogram 27 (h) The variogram is:
21(h) = c.'(1 ), + h)- Z(p,)) Vp,, .,,+ hE N
where 21(h) is the variance between samples a distance of ii apart from each other
and N is lhe neighborhood of known sample points.

3-1
The semivariogram can be represented by a formula or a graph. Graphs de-
pict the distance h on the abscissa and the semivariogram -y(h) on the ordinate.
The experimental or sample semivariogram (graph) is computed and ploted from
the known sample points and values and is compared against known model semi-
variograms to determine the best fit (closest match). After a fit is made, model
parameters are estimated. The semivariogram used in equations [81 and [11] is a
model function, not experimental.
Davis states that the following equation can be used for estimating the semi-
variogram (experimental) for multiples of Ii when h is the same between data points
(in other words the data is regular)
- "
S (x- x +h - (X - X h
ICY -X,..h). [13]
The asterisk indicates this semivariogram is experimental or estimated from
the sample values. This expression takes into account drift in the inner second term
in the numerator,
Z(X,-X.+h) [141
11
Equations [131 and [141 are in (14).

3- 15
Process of Structural Analysis
The goal of structural analysis is to determine a model semivariogram. To find
this model semivariogram, an experimental semivariogram is first estimated from
the known data values and compared to known model semivariograms to find a close
match. However, before this is done, it must be determined if drift exists. The
drift expression and the experimental semivariogram change based on the size of the
neighborhood 4 . It would be best to estimate the semivariogram using the entire data
set. However, it is too costly and often does little good since a distance is usually
reached at which the affect of values on one another becomes neglible. Therefore.
a maximum distance for the neighborhood is assumed initially to determine drift
experimentally. This same distance is used in calculating the experimental semivar-
iogram.
In the case of ordinary kriging, no drift exists so the experimental semivari-
ogram calculated from the orginal sample values is sufficient to determine the spatial
correlation of the samples. If drift exists, the situation is more complex because the
semivariogram is not reliable statistically.
The model semivariogram must provide good statistical properties, like corre-
lation between sample points based on spatial relationships. However. estimating
the semivariogram of non-stationary regionalized variable., may not have these kind
of properties. According to Davis. stationary variables (regionalized variables with
stationary means) force equation 1141 to zero. which gives equation [131 a known
statistical property - "the difference between the variance and the spatial autoco-
variance for the same distance." Normalizing the variables. i.e.. mean zero and
variance 1, provides an even better statistical property - the sernivariogram becomes
the -mirror image of the autocorrelation function" (14:2-12).
The main p~roblem in the presence of global drift is that the experimental
semivariogram is not reliable statistically. Recall that drift can exist in two forms.
local and global. Local drift is accou-ted for in universal kriging. Therefore. the
main task of structural analysis in this case is to compute a reliable experimental
semivariogram in the case of global drift. Since stationary regionalized variables are
considered reliable and residuals are stationary. then the r-siduals can be used to

4
The drift coefficients as well as the order of the drift polynomial may differ.

3I-l i
compute the experimental semivariogram. To find the residuals, which are the drifts
subtracted from the actual values, equation [91 must be solved.
The weights in equation [91 are determined in two possible ways. First, they
can be estimated by the Lagrange Multipliers in the universal kriging system or they
can be separately estimated by a regression technique like least-squares (13). If a
kriging system is used, a known semivariogram can be assumed at the start (a "first
approximation") to obtain estimates of the drift., If a least-squares approach is used,
a first- or second-order polynomial is fit to the sample data to obtain estimates of
the drift coefficients. In this case the di's in equation [9] are estimated as linear
combinations of the data
(17 = 5jjZ7,
21

After estimates of Ihe drift are obtained, they are then removed from the actual
data to obtain residuals. The residuals are then used to estimate an experimental
semivariograrn. The experimental semivariogram is then compared to known models.
If a poor fit results, the "first approximation* semivariogram, the neighborhood size.
and/or even the order k of the polynomial drift equation [121 can be modified to
obtain a closer fit, This recursive process is known as structural analysis. There is
a strong -interrelation between neighborhood size, drift, and semivariogram for the
residluals" (1-).
Once a model fit is obtained, a kriging system is used with the model semi-
variogram to obtain estimates fron the residuals. To get the final kriged estimates.
the drift estimates are added to the estimates determined by kriging the residual
regionalized variables.
In summaiy. the first step in structural analysis is to determine if global drift
exists inthe data set. This requires calculating a drift or sample mean for every
sample value in the data set. If it exists. global drift must be removed by calculating
resi(uals and then computing the experimental semivariogram from these residuals.
Once a model s%.nivariograim is chosen, values are estimated by kriging tlie residual
data set with this mnodel semivariogram. These kriged values are then added back
to the drifts calculated at the beginning of the analysis. Or. if global drift. does
not exist. the data itself is used to determine the experimental semivariogram. If
global drift exists or it does not. local drift can he assumed because if local drift does
not exist. the local drift coefficients will just be zero. Global drift is not so easily
accomnted for because it involves a much larger neighhorhood of sample values.

3-17
Models
Model semivariograms are used in the kriging system of equations rather than
experimental semivariograms. Experimental semivariograms are not used because
they do not provide results for distances other than those derived from the sam-
ple data. The term model refers to a known. continuous semivariogram (31) and
(14:246). An experimental semivariogram like equation !131 is not used in equa-
tions [8] and [..,] because the experimental semivariogram is compt:ted for known
discrete distances. Instead, a known model semivariogram computed for continuous
distances is used. Since kriging estimates new values at possibly different distances
than those between the original sample points, a continuous function is the most
reasonable choice (14:246). Some of the better known continuous semivariogram
models include the linear, spherical, and exponential (see figure 3.4).

armcce!

3! ~q
--oq
-,-i/- -:-"-e: - - - - - --. o

son'er ical mode!lnl I:-0e

Figure 3.A. Example semivariograms for linear, spheriral and exponent ional models

Once a model is selected. parameters of the model must be estimated. AC-


cording to (ressie H 1:198). Zimmerman and Zimmerman claim that tle "weighted
least squares approach usually performs ;,ell and never does poorly" at estimating
parameters for a model semivariogram from the original data points.
Two terms are used to describe the models. Since the data is assumed to
vary within some region, there usually exists some distance at which the data ceases
to vary and has some constant variance. This constant variance is a flat line onl a
semivariogram known as the sill, o'. For models with a sill, the sill is the sample
variance (6:7). The distance at which this constant variance is reached is termed tile
rangea, or zone of influence. The range then determines the neighborhood of sample
values used in the kriging equations because it represents the spatial zone in which
sample values influence each other. Beyond this zone, the influence of a sample value
on a kriged value is negligible. For a further description of sill and range see (10).
The following discussion of semivariogram models is based on (14:247), (31).
and (13).
Linear Model. The following equation is used for this model:
-y(h) = ah + b
The parameters are the distance h and the slope of the line. This model does
not have a sill, but sometimes the function is modified to create an artificial sill as
follows:
",(h) = ah + b for h < a
=a 2 for h >ei
IS

where a is the range. Davis suggests this approximation is good for "distances
much less than the range" (14:247).
.Spherical.lodrL The model equation is
L-!2h)
Q +Co if h<a

,(h)= C + Cj, if h>a


0 if h=0
The spherical model is most commonly used because it models most natural
data very well. The relationship modelled between data values is that spat ially closer
data values have a higher influence on each other than do those fort her apart - until
the range or limit of influence is reached. Once the range is reached. the values cease
to have a significant affect on each other. The model parameters are the distance
(h). the range (a). t lie sill (C + Co = ey'). and the nugg!lI rff,ci ((',).

:-
i!
The nugget effect measures micro-scale variations. It is the position on the
-1(h) axis where the semivariogram intersects, possibly causing a discontinuity at the
origin if Co # 0(35).

Exponential Model. The model equation is:


-y(h) = o -(l

The parameters are the distance (h), range (a) and sill (a2). This model is
characterized by a semivariogram approaching the sill asymptotically. This indicates
the data values always influence each other regardless of distance apart; however,
values sepamated by distances beyond the range have much less influence on each
other than those values separated by distances less than the range.

:3.6 Isotropy/Anisotropy

Matheron (35:1249) noted three 'qualitative characteristics" exhibited by re-


gionalized variables. The first is support. Support consists of the shape, orientation.
size. and spatial arrangement of the sample values. The second is that the sample
values within a support show some form of continuity. The third. termed anisotropy,
indicates the distance in the "zone of influence" (13:68) varies along different direc-
tions. Methods exist for treating different forms of anisotropy. but the discussion of
the forms is beyond the scope of this research. For furt her information on this topic
see (25) and (13).

3. 7 Chlptzr Summary

This chapter has described some of the most basic forms of kriging. Structural
analysis and kriging are complex. robust processes. It is optimal because the estima-
tion variance is minimized, the estimation is unbiased. and the covariance approxi-
mat ion, the semivariogram. analyzes sampie points based on their inter-dependence.
There are many current 3D medical imaging applicat ions in u.s today t hat can
use kriging to obtain more accurate estimates. In the next chapter. I will (lenonst rate
how kriging can )e used to estimate intra-cell scalar values in cell interpolation
surface extraction and the volume pre-processing operation of slice interpolatien.

'i-20
IV. Cell Subdivision and Slice Interpolation Implementation

4.1 Introduction

I implemented two methods to explore estimation in 3D medical imaging. First,


I implemc ed a cell subdivision method to investigate intra-cell scalar value esti-
mation. I also implemented a method of estimating values between two medical
data slices for the purpose of investigating estimation techniques in the volume pre-
processing operation of slice interpolation.

This chapter begins by presenting an overview of cell subdivision and surface


formation. Cell subdivision is discussed to explain how intra-cell points are derived.
Once values are estimated at, intra-cell points, the surface is formed. The process
of surface formation is briefly discussed in this chapter. Then, the two estimation
techniques, tricubic interpolation and kriging are discussed to show how they are
used to estimate values at the intra-cell points derived by cell subdivision. The
chapter ends by discussing my implementation of medical image slice interpolation.
Appendix C provided further implementation details of the cell subdivision method.

4.2 Overview of Cell Subdiri-'iovl and 3D Surface Formation


Presented in this section are the cell subdivision and 3D surface formation
processing steps. Each step is addressed in turn. Before listing these steps, some
terminology that helps in understanding the following explanations is discussed.
Voxels are treated as point values, not volumes. I consider major cells the initial
computational cells before subdivision (see figure 4.2). The cells created within a
major cell by cell subdivision. I 'crmi minor cells. Major cells have the original voxel
points and assc-iated values as vertices. I consider minor cell vertices as minor-
voxel points and values. Minor-voxel values are derived by intra-cell scalir value
estimation. Finally. I considei the arrays within a major subdiviled cell as mini-
slices, since logically they represent input data slices. The reason for this will be
explained in the discussion of st.eps 1 and 2.

The primary steps in the cell subdivision and surface formation algorithm are

1. Read data slices inlo nemory.,

4-1
2., March major cells between slices.
3. Subdivide major cells into minor cells.
4. Estimate minor-voxel values and normals.
5. Apply marching cubes surface extraction within major cells to
6. Render triangular mesh with a poiygonal based renderer. form surface.

Steps 1 and 2
These two steps are part of the marching cubes algorithm developed by Lorensen
and Cline (34)1. Four slices of data are processed at a time. The marching cubes
algorithm creates computational cells (cubes) between the two inner slices (see fig-
ure 4.1) and approximates the surface within each cell. My cell subdivision algorithm
works similarly, except, surface formation is approximated in minor cells, not major
cells2 .
Steps 3 and 4
When a major cell is processed, it is subdivided based on subdivision factors in
each of the three component directions. For simplicity, assume component subdivi-
sion factors are equal. A subdivision factor of two will divide a major cell into eight
minor cells (see figure 4.2). The 3D minor-voxel points are obtained by calculat-
ing division points along lines between major cell vertices based on the subdivision
factors. A subdivision factor of two creates one division point along each line, a
subdivision factor of three creates two division points along each line. etc., again
assuming the same factor in all three directions.
The scalar values at, these division points are estimated by one of the three es-
timation techniques - t rilinear interpolation, tricubic interpolation, or kriging. Tri-
linear interpolation assumes the voxel values vary linearly within a cell in all three
directions. This method assumes only the eight major cell vertices contribute to the
estimation of intra-cell scalar values (minor-voxel values). Since trilinear interpola-
tion 'is a standard method, details are not given here, but are in appendix H Tricul.uic
interpolation is a parametric cubic polynomial interpolation method. This method

'During this effort., I implemented the marching cube.5 algorithm described by Lorensen and
Cline (34). Details of this implementation are in appendices A and B. While developing the cell
subdivision code I re-used as ninch of the marching cubes code as possible.
2 IM%implementation can also perform the marching cubes algorithm on major cells.

.1-2
slice73

slice 2

slice 1

slice0

voxeli~j~kvoxel i+1Ij.k

Z voxel ij+l,k
voxel i+1,j+1,Kl+1

Figure 4.1. Computational cell (cube) marching between data slices

4-3
minor cell

Subdivision factor =2 in all dimensions

0 Original voxels, (major cell vertices)

0 Minor voxels (minor cell vertices)

Figure 4.2. Subdivided Major Cell


uses a much larger neighborhood of sample values to estimate minor-voxel values.
It uses the surrounding 64 voxel values, where the major cell is centered within the
next larger cube (see figure 4.3). Note, the 3D coordinates in the figure correspond
to summation indices in the tricubic interpolation implementation discussed in the
next section.

Y (2,2'2')

zy

I L

Figure -1.3. Major cell centered wit hin surrounding cube

Tricubir interpolation assumes values vary cubically within a major cell. i.e..
they fit a 3D) curved surface within the major cell. Trilinear interpolation and tricubic
interpolation are termed deterministic because the procedures do not account for
error. Krigiing on the other hand estimates values based onl the statistics of thle
data, and not, only accounts for error, but minimizes it. It estimates values using
a weighted linear corrlbination of nearb~y known samle values (voxel values). Thle(

4 -5
weights are determined by conditions that insure unbiased sampling and minimum
estimation error variance. The latter condition requires that co-variances between
sample values be computed. These co-variances are approximated by a technique
that computes the average difference squared (in distance) between data samples.
This causes sample values closer to the value being estimated to have more influence
in the estimation than sample values farther away. Therefore, kriging is really a
distance weighted estimation function. It does not assume linear,quadratic, or any
form of variation, although it can be tailored to do so. In fact, I tailor kriging to
behave like both a tricubic and a trilinear interpolator. Details of this tailoring and
tricubic interpolation are discussed in the next two major sections of this chapter.
The purposes of cell subdivision are to 1) resolve ambiguity in cells and 2)
provide a better surface approximation., Recall a cell is ambiguous if more than
one topology can be chosen to represent the surface within the cell. Cell subdivision
guarantees that major cells will be disambiguated only because they are being subdi-
vided. That is, once subdivided, major cells are no longer treated computationally,
hence they are disambiguated. The minor cells are now the computational units,
The problem with cell subdivision as a disamnbiuation method is that minor cells
may still be ambiguous. If minor cells are ambiguous, Wilhelms and Gelder (50)
apply the facial averaging technique discussed in chapter two to disambiguate Jhese.
I do not (1o this here. because my goal is to investigate the use of kriging to estimate
intra-cell values. Besides disambiguating amlbiguous major cells, cell subdivision also
provides a better surface approximation within each major cell. This is because the
surface is now being detected at a finer sampling - although many of the values are
not original Qample.s, but rather estimated values. How well the extracted surface
corresponds to the actual surface depends mainly on the accuracy of the estimation
function employed. To better underst and how subdivision can not only disainbiguate
most, riinor cells. but also form a smoother surface, see appendix I).

There are two marching cubes implementations use(d in the cell subdivision
process. The first, is the outermost loop. In this loop. data slices are read into nenory
and major cells are formed. At this point., a vanilla marching cubes imllelnentation
can be selected 3 . If the vanilla marching cc.bes im plententation is not .;elected. cell

aVanilla indicat es no cell suhd Ision. disambignalion, or enhancenients to the original algorithm

I6
subdivision is performed. The second marching cubes implementation occurs within
each nonempty major cell. In this case, mini-slices formed by the subdivision are
treated as if they are actual data slices. This minor marching cubes implementation
treats minor cells as "cdbes.' Within each minor cell, triangle vertices and normals
are computed to represent the portion of the surface passing through each cell.
Details of triangle formation and normal computation are in appendices AB and C.
Step 6
The triangular mesh is rendered using Phong reflectance and Phong shading.
Details of the renderer used are in Appendix F.
A more detailed discussion of cell subdivision is in appendix C.

4.3 Tricubic Interpolation


The tricubic interpolation estimation technique assumes that estimated data
within a major cell fits a cubic surface within the cell. In a two-dimensional sense,
this is like fitting points to a curved line between two known points (see figure 4.4)

Pl

Estimated values x
beitween known points P1 and P2

Figure 4.4. Example of fitting points to a curved line in 21)

The tricubic polynomial interpolation method implemented is tle satn"e one


iniplemented by Wilhelhs and Gelder (50:15). The parametric cubic polynomial
interpolation method they and I use is very similar to the hermite form of a cubic
polynomial curve, wit h constraints being Iwo endpoints and tangent vectors at the
endpoints (20). Two differences exist between the classical hermite form of a cubic
polyionial and the one used here. The classical form is used to generate point.'

.1-7
on a surface. Here we are estimating values that lie ol or near a surface in 3D.
not the points that generate the surface. Also the tangent vector constraints are
determined differently, based on the volume ,ata. The following equation represents
the formulation of the values in one dimension:
F(u)= au 3 + bu 2 + cu + d = U.,Cx = U. MH.GH = [u3 2
11 U IIMH - GH

where 0 < v < 1, MH and GH, are the Hermite basis matrix and geometry vector.
Normally, the tangent vector constraints are determined by differentiating the
U vector and solving at u = 0 and u = 1, the endpoints of the curve segment. 1,
as \Wilhelns and Gelder do, modify these constraints by assuming the derivative of
a value f, at a point i in one dimension is approximately the central difference fl =
l(f,+, - f;-I). The blending functions for a univariate curve F(u) are determined
by solving a system of equations including the constraints F(0) = fo, F'(0) = ] .
F(1) = fl, and F'(1) = .fl. The solution of the system results in the following
equation for an estimated value in one dimension
2
F(u) =EZfB,(u)
-1

where B,(v) are the blending functions

B = 1(- + 2 -,

Bo(u) =1 (3u _ 3
5u 2 + 2)
B() !(-3- +4 +
2
B2 (u) = ( 113_ )
The indexing scheme (-I to 2) is used to correspond to the position of a cell's

eight vertices in relation to the surrounding voxels. The point (0.0.0) is .lie bottom
front left. vertex of a major cell (see figure 4.3)'.

The tricubic function is determined by apl)lying the equation in all three corn-
ponents and is given by
2 '2 2

F(,,. , = .k) uj)(


k=-I.i=-i t=-)

'Although not shown in the figure, the major cell centered in the larger surrounding cube is
subdivided

1-8'
F(u, v. w) estimates minor-voxel values at the intra-cell points derived by cell
subdivision. u,v and w range between 0 and 1. u is equal to the fraction of tile
distance between the major cell vertices in the x direction. Similarly, v is the fraction
of the distance between the major cell vertices in the y direction, and i,corresponds
in like manner in the z direction.
This function constrains the surface to the computational cell because the two
endpoint values in the formulation for each dimension are cell vertex values. The
term tri stems from the three parameters. For the bicubic case, Watt (49) describes
the surface formulation as a cartesian product of two curves. In the tricubic case,
the surface formulation is the cartestian product. of three curves.

4.4 Krigitg Estination

This section discusses the kriging estimation technique used to estimate minor-
voxel values. First, an overview of the technique is presented. Following that, specific
implementation conditions are listed and discussed. Lastly, the kriging estimation
procedure implemented in this research is presented
The kriging code developed during this effort was written by Capt Chris Brod-
kin (3) and modified for use in intra-cell scalar value estimation. It. is object-oriented
code written in C++.

4.4.1 Overtchrw f Ariqing Techniquc Kriging estimates minor-voxel values


by a weighted linear sum of nearby original voxel values. The conditions of unbi-
aseduess and minimum estimation error variance yield a system of solvable equations
in the familar matrix equation form X = A-'B. where the solition N matrix con-
:ains the desired weights. However. before this matrix equation can be solved. the
model senivariogram in the A and B matrices must be determinedl.
The goal of this research is s~ifply to demonstrate that kriging can be used to
est imate ilit ra-cell scalar values and that it.
is flexible. To ensure that krigi ug can
estimnate points similar to the other methods, I tailored it to behave like tricubic and(
trilinear interpolations. To achieve this, structural analysis is not necessary. Ilow-
ever. st ructural analysi:, is necessary if the goal is to calculate tle best est inmates. To
tailor kriging to behave like tricubic and trilinear interpolations, some assumnpt.ions
W'*e lV
Made
* Global drift does not exist.
" Local drift may exist.
" the model semivariogram is known.
* The neighborhood size is known, but alterable.

" The data is isotropic.

4.4. 12 Global and Local Drift Drift is the phenomena that occurs when sample
means vary from point to point. A sample mean is derived by choosing some neigh-
borhood of values surrounding a known sample point and calculating a weighted
average. if these weighted averages (sample means) I;ffer from point to point, then
drift exists. If they are the same, then the sample .jeans are constant.
Drift can occur in two ways. It can occur within lo( I regions (local drift)
and/or throughout the entire data set (global drift). If global drift exists, it is
removed by calculating residuals from he .mle means. A residual is calculated
by subracting the estimated drift from the sample value at a I nown point. Residual
data is considered to be nonst.tionary, i.e., there is no drift. Since kriging only works
with nonstationary data, residuals are then kriged to estimate values. These values
are then added to the sample means to get the final estimates.
Drift can also occur just within the neighborhood of sample values being used
to determine the linear sum. This local drift can be accounted for in the universal
kriging system of equations.
Global drift. is assumed not to exist in this implementation to simplify the
process. Local drift is incorporated into the universal system by the polynomial
form of the drift. expression. Even if global drift exists, it should not have a large
effect if local drift is accounted for.
The three-dimensional local linear drift expression implemented for this effort
in terms of t he geometric coordinates T,. y,. :,is
in(p.) = in(x,,y.,Zi) = do + di x. + d2 y, + (13:,

and the three-dimensional local quadratic drift expression is


m(p,) = do + dix, + d 2y, + d 3 Z , + d4X 2, + d5 y.2
+drz + d7.,?yi + ds.rtz, + dylz,

•1-10
4.4.3 The Assumed Model Semivariogramnz The model semivariogram is a con-
tinuous function that takes the distance between two sample points as a parameter.
It is used in the kriging system of equations as an approximation to the co-variance
between sample values to give geometric meaning to the values instead of probabilis-
tic meaning as in classical statistics. The co-variance measures the inter-dependence
or correlation of sample values, whereas the semivariogram measures the spatial de-
pendence of sample values based on distance. Two types of semivariograms exist.
These are the experimental and the model. The experimental sernivariogram is a
discrete function derived from the data set prior to kriging. It is computed as an
average difference squared between data points. The experimental semivariogram is
then compared to continuous, known model semivariograms to find the best match.
Model semivariograms are actually used in the kriging equations because distances
other than those found in the data set might be used to estimate new points.

To tailor kriging to behave like tricubic interpolation, an experimental semi-


variogram was not computed and the model semivariogram was assumed to be a
cubic function. This assumed semivariogram model is also called a drift model, (31),
see figure 4.5.,

This model indicates a polynomial drift exists in the data, which is certainly
the assumption made in the tricubic interpolation method discussed above. The
function used is simply

(h) = abs(h 3 )

This model semivariogram coupled with the drift expressions above parallels
the tricubic method. The tricubic method is derived from paramet:ic cubic functions.

4.4.4 Neighborhood Size Recall from chapter three that neighborhood size is
tlie number of sample values in the kriging estimation equation. Neighborhood size
is usually determined during a structural analysis of the data (structural analysis
was discussed in chapter thre). Instead of dletrmining the neighborhood by a
structural analysis of the data, I assumed different sizes. First. I assumed the same
neiglborhood size as that used in tricubic interpolation. 6.1. This neighborhood size
is the 56 surrounding points. including the cells eight points, for a total of 6 sample
values in the kriging system (see figure .1.3). The neighborhood sizes 8. 16. and 32
were also investigated for bot h cell sullivision int ra-cell scalar value estimation and
slice interpolation. These sizes were chosen bwcause Ihey were easily implemented

-1 il
h

Figure 4.5. Drift model variograln

Ii2
I
from the cell geometry. They were used to demonstrate the effect of kriging with
different neighborhood sizes.

4.4.5 Estimation Procedure The goal of kriging restated in terms of this par-
ticular problem is to estimate
71

2(p) = ZWiz(m)
i=l

where the pi and Zi's are the surrounding voxel points and values and p is the 3D
point where the value Z is being estimated. n = 8, 16, 32 or 64 in this equation and
all the following ones for my implementation.
Both the ordinary and universal forms of kriging were implemented. Recall
the ordinary system is
n

,j+ n = 7-P
E--f' ( ,..,
3=1

This system of equations is represented in the matrix form AX =B,where


the X column vector contains the kriging weights.

-111 "112 -•-1 51VlWl

-'21 122 1
... ^1,2, 2 7)2p

11 .. 1 0 j1 i

=A-'B. where - is the inverse


The solution to this miatrix equation is X
miat-rix of A.

For die universal system, the equations change depending on tihe assumed local
drift, fl(p,). The system Is
n k

3=1 1=0

•1
-13
n )= P(p) (1= , ,. ,k)
F , a]-,fPt(p

The following equations contain quadratic drift. Linear drift is obtained by


removing all the quadratic terms. The universal kriging system with local quadratic
drift implemented in this research is
n
Z tW +, +Il X,+ , ,2Y+ 1-3Zz +,T44 rZ+ ,ISy, +?/.:I'
+ 7 -,Y + ,.1,
+ 11 Xz, + 119y :,
=

( srlVi=,.., '11 0)

= p
',IWz, ( - 2)

= ( = 3)
(=1=4)

(1 5

2
,
,
= . (l1= 6)
It.
Z'w,'z
E___

I
t=|

, (1 = )
,.rz, =.
,_

:, (I = 9)
!/,. and :,are the como)Oients of the i'th 31D sanipie poil.
',.

The following is the matrix form ol this s'stel.. AN H.

I I
711 712 . 7.
Yir 1 x] I Yl Z1 X1 y7 zi Xly xz y' z 1 U,
1 11P
2
'Y21 _f22 ... 112n X2 Y2 Z2 X2 Y22 2
Z'2 X2y2 X 2 Z2 7/2 Z2 IV2 72p

7nlI n2 )Ynn 1 Xn Yn Zn Xn Y 2 "n Xny2 XnZ,, YnZn Wnn

1 1 1 0 0 0 0 0 0 0 0 0 0 71o 1
XI x2 .. x, 0 0 0 00 0 0 0 0 XP
7Y2... Yn 0 0 0 0 0 00 0 0 0 712 yp
z z2 ... z 0 0 0 0 0 0 0 0 0 0 713 Zp
2 .2 00 0 0 0 0 j) 0 0 7)4 T2
1I 2n
2 2 2 2
Yl Y2 . Y 0 0 0 0 0 0 0 0 0 0 ?Is z
-.2 -.2 .. 2 A 2AA

x~yl x2 2y ... x yn 0 0 0 00 0 0 0 0 q17 xpyp


XlZ1 X 2 Z2 . .. XnZ n 0 0 0 0 0 0 0 0 0 0 18 xp -p

Yzl Y2Z 2 .. YnZn 0 0 0 0 0 0 0 0 0 0 19 YPz p

The A matrix is inverted to solve for the X column vector of unknown weights
and Lagrange multipliers. Recall the semivariogram 7,m is the same as
where hM,, is the distance from point m to point i. The 711's are the k + 1 La-
grange multipliers where k = 9 and (x,, y,, zp) is the point where the value is being
estimated.

4.,5 Slir. Ihtu'polation


..

This section descrilbes how scalar values are estimated between medical slices
for the volume pre-processing operrt;on of slice interpolation. To accomplish this
task I re-used as much of the cell subdivision code as possible. C'omlputational cells
are used it the estimation process. The estimation is dlone only along one cell edge
because dhe purpose here is to interpolate only in the 'Z' direction. Recall the '
direction is the direction tha, the data slices are stacked. Four different neighborhood
sizes were used - 16, 32 and 64. The slice inte polat.ioi algorithm is

1-15
1. Read in four medical data slices.
2. March cell (cube) between the two inner data slices.,
3. Estimate scalar values for new slice(s) along cell edges.
4. Create gray scale image from estimates.

Four slices are needed at a time because the tricubic interpolation requires a
neighborhood size of 64. Computational cells on the boundary of the data are treated
specially. Since they do not have access to the larger neighborhood of sample values,
I compute a linear interpolation along boundary cell edges. This should have no
effect on the final image because the boundaries of images typ;cally do not contain
any significant data.
My implementation of linear interpolation is not presented because it is a
standard method. The important point about linear interpolation is that :t assumes
only a linear variation. The tricubic interpolation and kriging estimation methods
used here are the same ones described in the previous sections.
The next chapter presents the results of implementing the methods discussed
in this chapter and in the appendices. The appendices contain descriptions of the
methods used to accomplish the other tasks outlined in chapter one. yet not men-
tioned in this chapter.

4-16
V. Results

This chapter presents the results obtained from implementing the estimation
methods discussed in the previous chapter. The results are divided into three major
areas:

" Artificial Volume - Cell interpolation surface extraction and intra-cell scalar
value estimation in an artificial volume.

* Medical Image Slice Interpolation.


" Medical Volume - Cell interpolation surface extraction and inti.:-cell scalar
value estimation in a medical volume.

These results demonstrate that kriging subsumes the deterministic methods


investigated. I modify the kriging systefn by changing the neighborhood size and
local drift assumption. By doing this, I show that kriging can be tailored to behave
like the other deterministic methods, i.e. it is flexible, and in some cases produce
images that look better than the other methods'.
Medical image slice interpolation is discussed prior to medical volumes because
some analysis was done on a subset of the medical volume slices.

Explanation of pictures and tables


The labels in the images in the pictures presented in this chapter are file names
that were chosen to be descriptive. For example. f2tricubic3.rle indicates that the
image represents the f2 function, derived by cell subdivision with tricubic interpola-
tion at a subdivision factor of 3. He indicates the Utah run length encoded format.
Similarly, f2krige3nodrift.rle indicates the f2 function derived by cell subdivision
with kriging estimation assuming no local drift, at. a subdivision factor of 3. Also,
f2krigenh321inear indicates the f2 function derived by cell subdixision with kriging
estimation assuming local linear drift at an assumed subdivision factor. I will leave
off the extension He or rle.Z ( .Z is compressed format) for brevity.

'When I discuss image accuracy or quality, this is m% opinion about the visual appearance of
the images.
The table headings are as follows. In the pictures derived by cell subdivision,
the number of triangles and non-empty minor cells are listed. These are denoted as
column headers "# TRI" and "# NEM" respectively. All other columns discussed
are applicable to all the tables. The first is "Images". This header indicates one
particular image in a picture. The header "values compared with" indicates another
image compared with the image listed under the column "Ilmages". The header
"largest est value diff" presents the largest, estimated value difference between the
two compared images. This entry provides knowledge of the extrene deviation of
sample values derived by two different estimations. As David states, "the most
natural way to compare two values ...is to consider their difference." I actually
take the absolute value for this column. David also states the a.verage difference is
an important measurement as well to understand the dissimilarity between values
(13). This measurement is presented in the column under the heading "avg diff of
values". The average difference of the values is derived by the formula
n

Zubsltmagei -value,-tmage2_vuue,)
n= n is the number of values estimated,
Iwhere.
imagel-value, is the i'th value estimated in the generation of the image under the
column "hmage" and image2_valtf, is the i'th value estimated in the generation of
the image under the column "values compared with". Finally, the last column in all
the tables indicates the average difterence percentage of the total scalar value range.
For example. the scalar value range for the [2 function in an artificial volume is 136.0.
Also, the scalar value range for the slice interpolation tests is 256. 0 through 255.,

5.1 Artificial volume

The main purpose of this section is to show that using kriging to estimate
intra-cell scalar values resolves an)iguity in (ells and that kriging is very flexible
compared to the other two interpolation methods. By flexibility I mean that I can
change parameters in the process that, alter how kriging estimates. This cannot be
done with the other techniques. Both trilinear and tricubic assume the neighborhood
and local data variation. The advantage of kriging is that it can be tailored to behave
lik., either of the other two interpolation method&, or it can be tailored to analyze
any size neighborhood and assume other local data variations besides linear and
cubic. First presented are images comparing kriging to the other two interpolation
techniques. fixing kriging to behave like tricUbic. Then, both the local drift and

5-2
the neighborhood size are changed to show how flexible kriging is and demonstrate
how important these two factors are. Many other factors can be modified to make
kriging more robust and more accurate, but the purpose of this research is to just
demonstrate its usefulness and applicability.
An artificial volume is created by calculating scalar values at voxel points. The
artificial volumes represent continuous 3D iso-surfaces derived from mathematical
functions. The details of this -rocess are in an appendix. The mathematical function
used is a hyperboloid with known ambiguous cases. The function, which Wilhelms
and Gelder called F2 (50), is
2
F 2(x. y. z) = 4(y - 1)2 + 2(a' - z) - 2(x + z - 3)2 + 1

5.1.1 Neighborhood size 64, subdivision factor = 4, local drift assumption


differs. Since the number of ways to compare images is quite large - taking into
account neighborhood size, drift, and subdivision factors - I began by assuming
a neighborhood size of 64 and a subdivision factor of four. Then I compared the
estimated values derived from universal kriging (with local and quadratic drift) to the
estimated values derived from ordinary kriging. The values were exactly the same in
all three cases. This indicates that at least for this data set and a neighborhood size
of 64, there is no local drift. I explored quadratic local drift in nearly all the cases
presented in this chapter, but found only minor differences between these images
and those assuming local linear drift. Therefore to reduce the number of images in
the study, i only present ;mages derived by kriging that assume local linear drift or
no local dri.

5.1.2 Neighborhood size 64. subdivision factors differ, no local drift assumed.
Next, I ran the program with the three different estimation methods at the sub-
division factors two, three, four and five; still maintaining the kriging neighborhood
size at 64 and assuming no local drift. The results can be seen in figures 5.1 - 5.7.
All the cell subdivision methods - trilinear and tricubic interpolation and kriging -
removed the ambiguous cells for all subdivision factors. In other words, after cell
subdivision and intra-cell scalar value estimation, no ambiguous cases were found.
The subdivision factor as well as the estimation function significantly alters how the
surface is represented. The higher the subdivision factor, the smoother the surface
representation. Also, tricubic interpolat ion and kriging tailored to behave like tricu-

5-3
l1.T Miterpoi)
'Io t u )";: v-1. 1ilahtValdE" L1hat iq~tc 'mui ll . ': ~-'I ill iotI '

Fiure ). \ndl~a mnarch)in, (Ube, etIt lct iOt


01 f I]\
; rboloid :cIil face

Cp.11 4 alar i dne e'!t in1Z1 loll look~ aliiiust 1(101 iral. IThis i Is !( it a oIlilarisonl of
01 , t I I,! t cdf \-,I II, bet wc I I I lie( two) 1tletlleJd'. F, 'r* o\e IlI)p]', at 'Iii tI\ i:sion factor
-.val des et lilatc w.1triclibi)Intterpolat ioll adl dffe'r b\~ ;III averageZ of .011:3
ramwi~ of stalm \ alw
u Is -.19.0 ito 87.0. .Sov table, 5. fo let dli ced ( oruparison of the
('st imat loi (' li(IIII(i'. T1Hf IlIdjcatc." triang''le md~cNENI inc'ari Nuiiemptv Minor

ccV~
!lc'"111' arc, pnrecItc'( In tisllalee. BY ai iai hii, tabile and exanunl-

tig i, illat-'s :t i" 'Icells that I he krigiuit! a.'silillt icon, mlodel t IcI-lim Interipolation
iost'l\. Fo.r .lcIilo fact-rs 2. ;111( 1. 1,liv Iiiie of tI iniles a., \vell a-i the
tiiiiu1bl, of Iloivflpc'u j)!:1\ lllo CvIlk ;it'( exact l- 1'.I 1au1 for lilt tt ice and kriging
e."pmt.
111'I F'il. Indicate;, that theW c'ttimacd \al'r at'' \e * Sitmilar'. To
l':
firtIl- lc, tilt-. I cmtparIel vaiities (''ctitnlated as5 'i)wtil ini Ow tabile. noting the
Iat~c, ciff:(iet'(1 III Vxc
\alii and t'111t.c tffr'i. f.l Oit ('stIlrtiate(I Values.
ii''~"i-11- ce'; t m;Ip
h e it'tr . t11.iclev l iwj-iccvhod. tIC) ciift.
able Comparison of values estimated by trilinear and tricubic interpolation

and kriging (with a 64 neighborhood, no local drift)


values largest avg avg (liff
# # compared est dlift
Image TRI NEM with value of 136
diff" values 7
f2v-anmc 144 I -

f2thrilnear2 140 70
f2tricubic2 140 70 f2krige2 .136:3 .0031 .002:3
f2krige2 140 70
f2trilinear3 332 166
f2tricubic3 :348 174 f2krige3 .1211 .0069 .0050
f2krige3 3-18 174
f2t rilinearl 560 2S0
f2t-ricubic4 576 288 f2krie4 .1331 .011:3 .0083
f2krige4 576 288
f2trilinear5 888 444
f2t ricublic5 920 460 f2krige5 .1316 .0160 .0118
f2krige5 912 4.56 _

5-:
Figure 5.2. Cell subdivision.factor 2. with trilinear interpolation estimating minor-
voxel vaihes and nmarching cubes extraction of hyperboloid surface from
mm i-slices

h3 model senivariogram. and isotropic data) model tricubic interpolation extremely


well. However. as tli next inlage shows.,a iteiglborhood size of 6-1 i-not necessarv
f-" krigMing to estit.dle values to the M-pposed accurate value.- of tricubic in-
t..e
Lerpolation. Reducing the number Of samrple value- directly affects execution timle
because it reduces the size of the matrix being inverted in the krizing system of
equations.

5. 1.3 .\'tiyhhitlhnwI
dir Sbirts.oin faci
,iifo,. = 2. lochl drift t..sumip-
lion dfJter.. The next set of iniages in Lhis sect ion deinonstrites tlie flexibility of
kriging (figures 5.9 --.13)., In these .ets of irage. the subdlivisioin fa,-or is two.
chosen arbitrarily. I altered both the neiglhorhood size and the local drift. In all
tile imnages except -hose wili a neighborhood of 3"2. ihe itnages wt it) no drift are
significantly different than those witi local linear drift. It. appears dia. with smaller
neighborhoods. assuming local drift prevents the inaccuracies seen by the holes.

The neighborhoods selected are as folhws (see figure 5.S). A 32 neighborhood


starts withi the-oriinal 64 values stirromnditig the coniputatimonal cell and ignores the

"5-1,
Figure.5.3. Cell subdivision. factor 2. with t rieubic interpolation estimating minor-
voxel valhte. and marchins cubes extraction of hyperbolol'i surface from
win i Iices
-s

values in the 2 sli'es above and below the computational cell. A negilborhood of
16 in the "N",iheci iol,. denuoted nli l x in Lhe figures. coisist.s of the ,ight compu-
tat in. crll vorti ces anl for out pither side of the cell in the X ,li:,ection. 16 'Y'
and 16 Z neiglhlorhoods are d,.rived in a similar fashion. These particular neigh-
borh-)ows were chosen berauti of their direct correspondence to the :omputational
cell and Ihe indexing scherie already usefd for tricubic interpolation and kriging in a
64 twighborhooI.

uuage. delic'l aug difference,; wert created by the Utah R L libra-, .tool rlecomp
witi ohi,' ofOoperator. This tool performs Ihe' logical set differei'e Ol)e:'at toll between
pixel vales it) the two inmaiges.
Table .5.2 dlrict.- differences in rsihmatd values for the last -ets of images.
Notie that kriging with a neighiborhoorh of :2 sample valu.sc and ,I, local drift
matche. tric tbir interpolation almost as close.ly as krigitn with a ne-i.hborhood of
64 sample values does. I compared local linear drift kri-,ing images 1o IrilIbir and
linear , caus- local lina-ar drift kriging prodi-,,d 61etter iniage. \: soon as the
ktiginum neighborhool sie re-duci's to 1,. except in the ca.e o' nhliti . the kriging

1.if
Figure .5.4. Cell subdivision, factor 2, with krigiiig estimating minor-voxel values
and marching cubes extract ion of hyperboloid surface from mini-slices.
Kriging uses a ne'ighborhood of 64 ,arnple values and assumnes no local
drift.

5-8
Table 5.2. Comparison of values estimated by different kriging forms for cell in-
terpolation surface extraction of a hyperboloid surface in anl artificial
volume
J values largest av-g avgf diff
I compared j est duff
Image With value of 136
______________V diff
__________ values %
f2krigen h 32nod rift f2j, :?nh32inear .0310 .0004 .0003
f2krigen h32nod rift t2tricu bic2 .5066 .0098 .00 72
f2krigenhi 6x nod rift f2krjgenh16 xlinear .1922 .0023 .0017
f2krigenllxlinear f2tirictihic2 :31.7-500 1.2602 .9266
f2krigenh 16linear f2t rilinvaar2 .0646 .0006 .0004
fMkrigcnh I6ynodrift f2krigenh I 6ylinear .161-7 .0022 .0016
f2krigenh 16N6liiiear f:2tricubic2 A.0790 I.0210 .01.51
f2k rigen 1i163l inar f2trnh near2) 2.0-,,,0 .0-118 0307
f2k rigenh I6znod rift f2krigenhi I6zli niear .1922 .0023 .0017
f2krigenh IGzlinear f2tricubic2 :31.7-500 1.2602 .9266
f2krigenhl6zlinear f2trilinear2 .0646 '0006 .000.1
f2krigeiih8nod rift fMkrigenhli lnear .9313 .01 :38 .0100
f2krigenh~finear r2t.ricilmc2 .11.7-500 '.26i02 .9266
f2krigenh~liuear f2t.r'invar2 .0616 .0006k001

~~ ....
... ~ 5 9-----
Fi~tntre tiudivision fartor :3. Cpper left. vanilla marching culpt-. I pper right.
triii~ln:r itirrpolation. 1Lowf-r left. rictibic Inlterpola~tion1. Lower right.
kriginla wih 64j nielibhorhoxxl. neo rift

nIticl cler' to t rilinetar


1oai littar itruag, S111,.Lmch ilterpwilt ion. 17k? her t hani tricthic.

h.rtx;~pl... inige f2ki


kI e'iiIhlbiinear omipar.'d to imuage f.trifine.ir2 prothice, flip
Sli, l av
iv dit a iffertice pecinae vet i)O) for kri-61m. utii;,tes COMiParedl to
Ifi' tr tw inteplbrxatio lilth,)ds. *rkindicates that krisinm %,.ift aighoroo
of Ii; in direct ions *V or *V. local iinear drift and the modelI semi-variogran1 IrP
e!$t ilialteS tmlor VO:CI ViiaWS 11l111.t p~art.lv as trilincar lepiatm The resuilts
(leriveli fromn ui.mug krigirag in a neihborhio of S art- :uzilar itt qu~ality to t hme
derivedl ,viei-gllibrhlt)0s f6 *N itud *V. Tillm, indicate that
by Lialt lrvist Ibr Otis artificial
Voijlimti lihit smI, Ole addlitfinal %aitetsIl iltt l 1N and lb It, N iwixrlicmcds add
viri tadi not hing of signifcative to t he tnt niates.

H119rall Ihat I was att Iemptiniv to tailor krliim, to modeel iti ti bic initerpolation.
Fiowever. a- ttilt-:.able and iftiages indicatc, withlotit chaigig any pair.'ttrseXcept
neighbi~rl ize
i. If appears that
'nrigilig, list) l).laves likv t rilitivar int erjplat 11)1
krivin-g !:,.ec withIt his pa~itIicilar dlat a is morr itifliwiicerd IPY rieiStibrhood size and
local1 drift asipStjtions ratllt'r t han the sei:kario-arain itiodel. *I' te-t tis frterw. I
tletl~dwit l,
1".~wr seiiiva ritoggal II mdels of hi and iP-*, I phbta.itifd ;ivrraqr diffkrvilces
Figue- 5.6. Subldivision factor 41.Ufpper left. vanI~1a marching caaws. Upper right.
rilirar
titepol io I cwer left. :rrithic interpoiat ion. Lower riHt.
krivitug vit h 6i-1 neighborhood. not ririf:

of esini -tell valu:es le~is thatn .00051_. The uiitta wms romparmd if) data estiinatrd bY
the h ni,1~ . holding~ all Ot her ;ssu:mip:i it. he . Zame.

The 11tist atc:Irlitt Ithohd is tlitficnilt its blerertuanIeh~i1estimated val-


Of's art, .4. rc,, and beealti' later imiates uimy e=atter genterated medi11cal data
indica e Ithat Lriczibic and t he correspodinitg kri'gi:g me? h-x vit it t riciihk like as-
stinptsonst are not very accurate. For artificial viinmes. such a., the one containing
this hvuerbteloirl -murffce. tlit-Nno hems lokisig iniadge, are genierated b i he cell sub-
di%itu'it oil~sini. Iricithir ittte. pniationt oi- krig:ng at higher ,ghisinfat tors.
The problem with considering~ only art :twtai volitmes INt hat shr do not ad-
"puauely rejirem-1ns Ihe wnrld of titraical lit pai . rular. Ihere i. tio- noise or
sharp cont rasts lint e vatiur%. The next sert ifinl~ores how es.t imaiion techniques
performi in itividical ituaqr sliee i tterpolation.
Fiouare 5.7. Suibdivi~sioni fact:or i. Upjwr~ Vlet. vanilla marching che.Upper right.~
trilnear iztiterpiwatizn. L.ower let. tricithic inierliolaion. Lower right.
kriijn--t %vithi 641 n:wibhrlotul. no drifr

.,....f11fly 4lrdcI
1ant'.r
1uhlt tjiflailf

MIirlitctl inta~f. -5!we i III


o~aionisg a scn tra::sfortnatir,. In thin !,ectiou.
2D iniagae cata derive,! fromi M\ill am. CT :t titIit isI ransfiirni fr.nt ',y the slice
interpolation. SlihI interpolation create, logical slices letwtnexe:b Iv ones by esti.

mat in., srwalar values in the TZ dimension. In this- sectioni. I dem~on!,: rate1 lte use o
linear interpolationi. !rictic interpt$)ati;)n. and kr:ittng I.s Lit# interpo'alion methods
used ii es1; ;1ilat;. daAValues. Tritinear atnd linear are equivairtitt in thi%cawe because
I iterimiat lintl )IIoe rlitir~~iot. X. alzuu! a cell edge.

5.2.1 Dnq hrnurt. CT. The fir-o~nv(Oii.~ of 2V.2 N 13Q, pixel CT slices
ofa dto&*, heart. we pirit tre in figure 5.i4. *1h~e govl is to create.a new slice between
slires -111anti -12. The top fonr iniage! in IIhe pirturc are the original data slices.
The remnaining isnawes are attettilu, to estiniale a logical data --live betwern lte two

'Rrraij t1j~s se-lr spare is 11w t~lt -. rt~ed fre,, wttu nal
uvnx sm4lahtit in !letman and
Computational cell being

nhl32 -

val ues A

Computational cell being


p~rocessed

11116X - 16 samp~le
values in X nm 16y - 16 samnple
directioni

zA

values inl X

0 Sample values
contributing to estimation

Figure 5.8. lKriging neighborhoods, nh32. nllk, nh,163 and nhlft


Figure .5.9. INriging estimation. subdivision factor 2. neighborhood .32. Upper left,
no drift. Upper right, local linear drift. Lower left. image difference
of upper right from upper left. Lower right, image dlifferenice of upper
igh-t iigefrom fbtxicibic2.
Figure .510 riging estirnation. ,ubdivision factor 2. neighborhood 16x. Upper
left. no drift. Uppe riht. local linear drift. Lower left. image differ-
ence of upper right fromn upper left. Lower right, imiage dlifference of
upp)Ier righit imiage from fMrilineai-2.
Figure .5.1 1. Nriging, estimation. subdivision factor 2. neighborhood 16y. Upper
left, no drift. Upper right, local linear drift. Lower left. image differ-
ence10 of upper right from uipper left . Lower right, Image difference of
upper right image from P2triciibic2.

5-16i
Figure .5.12. I'riging estimal ion. subdivision factor 2. neighborhood l6z. Upper
left. no drift. U'pper right, local linear drift. Lower left. image differ-
eiice of' upper righit from upper left. Lowei right. image difference of
tuppt-r right i luage from fMrilineair2.

5-17
U UJ

Figure .5.1:3. Krigin'g estimation, slibcldhso factor 2. neighborhood S. Fpper left,


no drift. Upper right. local linear drift. Lower left, image difference
of upper right f-rm uIper left. Lower right, image difference of upper
right 'image from f2trilIineai-2.
mlidldle images in thle top) row, using linear and tricubic interpolations andl kriging
(in various forms). Table 5.:3 presents results pertaining to these images.
Notice in the picture the imiage genierated by tricubic Interp~olationi is inaccu-
rate. The inia-ees produced by krirviii wit neighborhood 6-1 are also inaccurate and
according to the comparison in the table. estimate values similar to tricubic interpo-
I-tion. Thiese inaccuracies are possibly caused by twvo reasons. First, thle estimation
functions might he accounting for values that should not be inifluencing the esti-
mat ion. Second, the assumption of cubic variation might be invalid. Nothing canl
be changed with the trilcubic interpolation function to prevent these inaccuracies.
However. parameters In the kriging process canl be modified to p~revent them. These
paraineters are neighborhood size. dIrift assumptions. and the semivariogram. Ini this
study. I ol\ r*'odify the drift assum-ptions andl thle neighborhood size.
Notice that excep~t for kriging with neighborhood 16 'Z', khriging with smaller
neighborhoods estimates values closer to trilinear than to triculbic. and gets thle
closest match to trihinear wvith a neighborhood of S. assuiming local linear drift.
Ordinary kriging with smaller neighborhood sizes of 16 and S produces significant
errors iii the estimated values. However, universal krigingE (local drift assumed)
corrects these errors.

.5.2.2 Biubij liad. M1RI. The second 21D medical imagre st mldv (see fiagie~ -5.15
consists of 166 X 166iN1111 slices of' a I hree nioint 1hold babi's head. lnterslice thickness
is 41inni. The goal is to estimate a n~ew slice between slices :31 and :12. Thie original
slices again are Ii I lie top row and( thle secondl row shows slice, are dlerived by' linear
and trictibic interp~olationi. All others are dlifferenlt forms of kriging.

Alt hon-gh all the est irniaed imiages (except krigenhinodrift ) look almost iden-
icaId. anl e.Nariiat ionl of (lhe estimiat ed values reveals that estiniat ions in t Iiis stuidy
are similar to those Ii the (log heart study (see table 5.4).

..3 1I( dical vo/umn( cel/ inkpl


C7Oation $1ifa Cxa icion

Thle finial Series of imlages con .:st of it :31) sin face extracted from 60(iu)I:man
:)alv\ head INIM data slices. The 2D)slc dimiensions are( 128 p~ixels X 128 piNels
and thle ,iso-value ,is 413. Thel( surface met hodls I utse are the vanilla marching cubes
(fign're -5.1(i) and1( C(]l subdivision. I .,i- lhe cellI subdivision i echnique. I use a sild(i-
Visiu :i fact or of 2 in ihe XZ (liiiiel~oi and I inl bothl Ihle XN anid VY. This has flhc
Table 5.3, Comparison of dog heart CT estimated values.
1 values largest avg avg diff
compared est diff ---
Image with value of 256
ciff values %
krigenh6. ',,odrift
1 krigenh64linear .6229 .0196 .0077
krigenh6-1linear tricubic 4 1242 .1365 .0533
krigenh64.inear linear 18.0780 .6781 .2649
krigenh32nodrift krigenh321inear .9815 .0416 .0163
krigenh321inear tricubic 12.1747 .5519 .2156
krigenh321inear linear 3.2616 .1295 .0506
krigenhl6xnodrift krigenhl6xlinear 30.0089 1.3080 .5109
krigenhl6xlinear tricubic 12.4780 .5407 ,2112
krigenhl(i linear linear 2.27,19 .0785 .0307
krigenhI6ynodrift krigenh16ylinear 70.8097 .9918 .3870
krigenhl6ylinear tricubic 12.7687 .5527 .2159
kaigenhl6ylinear linear 3.0524 .1129 .0441
krigenhltznodrift krigenhli6zlinear 24 5295 .4502 .1759
krigenhl(izlinear tricubic 2.7525 .1125 .0439
krigenhl6zlinear linear 17.1736 .6742 .2634
krigenhnSodrift krigenli8linear 203.168.5 10.8014 4.2190
krigenhSlinear tricubic 1-1.4007 .5639 .2203
krigenh8linear linear .292:3] .0091 .0036

5-20
Table 5.4. Comparision of baby head MRI estimated values
values largest ax,, avg diff
compared est. diff -
linage with value of 256
diff values %
krigenh64nodrift krigenh641inear 1.8551 .0571 .0223
krigenh641inear tricubic 6.7329 .3772 .1473
krigenh641inear linear 23.0072 1.-960 .5844
krigenh32nodrift krigenh321inear 3.4845 .0932 .0364
krigenh321inear tricubic 17.1510 1.0481 .4094
krigenh321inear linear 5.1335 .2807 .1097
krigenhl6xnodrift krigenhl6xlinear 96.5233 3.4776 1.3580
krigenhl6xlinear tricubic 17.3192 1.0851 .4239
krigenh 16xlinear linear 4.27:38 .2252 .0880
krigenIhl6ynodrift k rigenhl6ylinear 70.1153 2.-"202 .9845
krigenhl 6ylinear tricubir 17.4163 1.085:3 .4239
krigenh 16ylinear linear 5.2594 .2249 .0878
krigenhl6znodrift. krigenhl6zlinear 21.8722 1.0342 .4040
krigenh I6zlinear tricubic 4.1201 .2519 .0984
krigenh 16zlnear linear 21.4664 1.4106 .5510
krigenlinodrift krigenh8linear 227.1227 16.3677 6.3940
krigenh8linear tricubic 17.9036 1.1631 .454:3
krigenh81inear linear .7859 .0282 .0110

.5-21
WI Vmer S,,lw(' l ;Ili, 12. \\it'o t Ic'- fiVt'jCI 'Itt ~c
e1.d 'slit ion

effectr )f pe'frIn g Ihe ' olnuIII I, (- -IiC )T.li! OIMerdt-ill kd slice Ifll-,r )d1i h Ifl. Tll(

Vi mode ''iuxdn)rl ., all be "certil I 1114 pictw-vc, !htmi 5~.17. 18. and
.5. 19) iiliclrt tilit f f'r-WhlhU nIl er-pulat ionl inl ihe Vol illm-t In~-
oCC!'lIM!
0Oper1al'it n >-4im ii ant ly affect I lwie ial rc' ostriIIcledlsnfc~ TrIlinear lilt erpulatioii

look 'liim llar1 td~


eac lici . I 1o%\eveu. diffelt're uib.~' et xxeeu the
Iili Iidicate

Owhn. are '.uglliticallI dilcrence. toi ev~imijlc. ligin 5.. 2(1 b'okv, - litv pI~vl d~icfi cet
bIt Wocll th It( idges11': gell alc h%.i
tVillo~ I I ihiita.'ijttvi-pot) I'mtl antd ku u\?

t intii; -.:t l llt Ics. t ri-ili 1c lilt jtdt ItII. Iulinear int i at itni alitIIr!ii±
os

pre-,W~~~~~~~~~~~~~~~.
wIbdl I' ,l ~c I ;1 al ~io-tdIl;
FJQUUI
,. * ,.
Bab
. il.1o-ail1 ".C l il-i ;I n
X by 4 by 4 Z volume. The ambiguous cell cases in the volume were removed in
all cases by cell subdivision, regardless of the estimation technique employed or the
subdivision factor. These images demonstrated that kriging is very flexible. That
is, modifying the neighborhood size and assumed local drift significantly alters the
waN kriging estimates values. Larger neighborhood sizes of 64 and 32 cause kriging
to behave like tricubic interpolation, regardless of the assumed local drift.. However.
at neighborhood sizes of 16 X', 16 'Z', and 8, kriging with local linear drift behaves
almost exactly as trilinear interpolation. The use of universal kriging was essential
at these lower neighborhood sizes for correcting inaccuracies produced friom using
ordinary kriging. The best appearing images in the artificial volume occurred with
tricubic interpolation and kriging tailored to behave like tricubic interpo1 'ion.
Different results were obtained by estimating 2D medical data slices between
two existing ones. At neighborhoods of 16 'X', 16 'Y' and 8, kriging estimated more
like linear interpolation than tricubic interpolation. Again, the use of universal krig-
ing was critical for correcting gross inaccuracies produced by using ordinary kriging.
The most significant result in these studies is that tricubic and kriging tailored to
behave like tricubic, estimates values poorly. Little can be done to tricubic inter-
polation to fix these inaccurate estimations. It inherently assumes a cubic variation
and uses 64 sample values for an estimate. However, kriging can be modified to
overcome these' inaccuracies. I did this by i educing the neighborhood size. The best
image in the study was obtained with a neighborhood size of 8 and local !mnear drift.
This image may not represent the best est imate(] values however. The best esltimate
can only be derived by doing a structural analysis of the data to determine if the
data is isotropic. to find a possibly better model semivariogram, and t.o determine
the optimal number of sample values., Furthermore, the optimal neighborhood size
may not be 8, 16. 32 or 64.
Finally, I presented a study of images derived from 60 MRI slices reconst ructed
into a 3D surface representation. This study demonstrated that tricubic interpolatioin
inaccurately estimates intra-cell scalar values for the purpose of cell subdivision
surface extraction - at least for the data set I analyzed. Both trilinear and kriging.
modified to behave more like trilinear than tricubic interpolation. produced much
better appearing images.
Showing that kriging can estimate like standard deterministic methods is it-
port ant. It shows that just guessing the kriging paraimeters produces results as well

5-26
as those already used in practice. This means that by using kriging, the estimation is
no worse than the standard deterministic methods. However, kriging theory states
that kriging will produce the best estimates if properly applied. Properly means
performing a structural analysis of the data, to determine a model semivariogram
that models the spatial correlation of sample values.

Since kriging subsumes the other deterministic methods investigated, kriging


alone could be used if different interpolation methods were desired for one rendition.
This dynamic choice of an estimation technique would be most useful for interac-
tive applications. where different estimation techniques might be useful at different
resolutions. For example, consider the baby face skin rendition shown in this chap-
ter. Trilinear interpolation or kriging tailored to behave like trilinear both appear to
be good estimation techniques for this particular viewpoint and subdivision factor.
However, suppose a radiologist is interested in viewing just the nasal area. Trilin-
ear interpolation with this subdivision factor may not provide an accurate enough
estimation for the detail required for a close up examination of this region. The
subdivision factors could be adjusted as well as kriging parameters such as neigh-
borhood size. semivariogram and local drift assumptions. to provide more accurate
estimations for closer viewing.

5-27
VI. Recommendations and Conclusion

This chapter first discusses recommended research using the 3D imaging tech-
niques implemented. Fo1':wing that, future kriging research applied to 3D imaging
is suggested. Finally, a brief conclusion to the thesis is presented.

6.1 3D Imaging Recomnendations


The 3D imaging methods implemented in this work can be used in the fol-
lowing ways to aid future research. First. since the marching cubes implementation
processes cells, it can be used as a foundation for any o r the cell-by-cell processing
techniques, e.g.. (46), (50). and (32). Although the implementation currently main-
tains only four slices in memory at once, it can be modified to maintain all or any
portion of the volume in memory simultaneously.
Another topic is different approaches to cell disambiguation. The simplistic
method of facial averaging in the soft object. method (52) can be modified by apply-
ing different estimation techniques to estimate the center point of ambiguous faces.
This was the focus of Wilhehn and Gelder's work (50). The software implemented in
this research does not resolve ambiguous cells. Subdividing alone does not guarantee
disambiguation like facial averaging does. The reason for this is after subdividing an
ambiguous major cell. minor cells calI be anibignous. To obt ain a smoot her looking
surface as well as resolving ambiguous cell.,,. t lie best met hod would be to subdi\ ide
and apply a method like facial averaging to completely (lisanl)iguate minor cells. I
suggest that anY future research begin by implementing the facial averaging tech-
nique applied to majoi and minor a miguous cells. Following that. lhe gradient con-
,istencv heuristics should be impnlmented for comparison. Finally, kriging should be
invest igated as a facial center value esi imat ion method. Since Ilv tessellate rout ine
is shared bet ween vanilla marching cubes and cell subdivision. est iniat ing anl)iguous
center face values and disambiguat ing can be applied easily to bot Ii majol and minlor
cells.
If a fast surfac, extractor is desired. several changes must be iade to the
implementation. fhe main bottleneck in the system is the I/0 overhead incurred
while writing and accesing geomelry files on disk. Storing points and norrnials in a
memory list and Ithen passing them straight to a ren(lerer should drastically cu tile

6 I
processing time. Also, utilizing a hardware renderer should decrease the processing
time even more.

The cell subdivision code could also be re-structured to operate in parallel.


Since only four data slices are ever anal'yzed at a time, the parallelization could be
partitioned by sets of data slices. Each set could be processed oil a different processor
to generate a portion of the iso-surface mesh. Rendering could also be partitioned
among processors by the same logic.

6.2 Krigitng Recommendations

The kriging estimation methods implemented in this work can be modified and
enhanced to investigate further uses in 3D imaging. The results depicted and de-
scribed in this thesis indicate scalar value estimation needs to be investigated further.
Tricubic interpolation does not estimate values accurately in the medical data sets I
analyzed. This study demonstrated that kriging is very flexible and can be modified
to behave like different estimation methods, including both tricubic interpolation and
trilinear interpolation. A research effort should be conducted to determine how to
make kriging find the best estimation as the theory indicates it should. This requires
a structural analysis of the data to determine the characteristics of the regionalized
variables - support, continuity, and anisotropy. Continuity of sample values exist
in certain portions of the human body such as organ and bone. \What. needs to be
determined is how to find these zones of influence. Finding the zone of influence
determines the kriging neighborhood and the type of model semivariogram to use.
In addition to the model semivariogram. I also made assumptions about the drift.
the neighborhood, and isotropy. In some cases. such as a neighborhood of 64. the
data might be anisotropic. If assumption of isot Vopy or the model semivariogram was
wrong. fixing them according to a struct ural analysis should make kriging produce
.,optimal" estimates.

Before structural analysis can be done however, certain tools have to be built
or tnodified from existing ones. Several structural analysis lools have been built at
AFIT for use ii 2D data sets. These would have to be modified for 3D use. The
tools include one that calculates the experimental variogram. It currently e stimates
the semivariogram in only two directions (to check for anisotropy). This would have
to be modified to estimate semivariograms in other directions for 3D. A procedure
exists that deterinites para'neters for a seinivariograin model. There are only a few

6-2
models available in the tool. so model implementation is another area mf research as
well as modifying the existing ones for use in volume data sets.

6.3 Conclusion
This research investigated different methods for estimating scalar values within
computational cells and in the volume pre-processing operation of slice interpolat ion,
These methods include the deteiministic linear. trilinear and tricubic interpolations
and the geostatistical estimation technique, kriging. Iso-surfaces were generated
by marching cubes and another (ell interpolation method called cell subdivision.
The estimation techniques were used to estimate intra-cell scalar values in the cell
subdivision method. They were also used to estimate logical data slices between
existing ones for the volume pre-processing operation of slice interpolation. This
research introduced kriging as an estimation technique for use in 3D imaging.
I demonstrated that kriging estimates values as accurately as deterministic
tricubic interpolation - shown to be very accurate in estimating intra-cell scalar
values in artificial volumes. I also showed that tricubic interpolation can perform
poorly in medical data sets and that kriging can be modified so these inaccuracies
do not occur.,
The erroneous results produced by t.ricubic and several of the kriging variat ions
could be caused bv invalid assumptions about the neighborhood influencing the es-
tilnation and the variation between sample values. Neither of these factor, can be
modified in tricubic interpolation: however. they can be in kriging. I only modified
the neighborhood size and local drift assumptions. These modifications demons' ra',e
lhat kriging produces better results than tricubic interpolation. The data variat ion.
which ,is
modelled by both the semivariogram and t lie local drift, needs to be deter-
inined )va t ru( tural analysis of tle data. MY goal was to demonst rate that kriging
is capable of being modified to behave like other (leterministic interpolation tech-
niques and to prevent inaccuracies. Since I did not do a structural analys;: and(just
assumed the data variation, the results show that, kriging is very robust. It is robust
because even though I guessed several of the kriging parameters. I was able to make
kriging behave like three other standard deterministic functions and etter in some
cases. Following a structural analysis, kriging should provide the best, estimation in
comparison to olher known est.imation methods.

6-3
The abiiity to modify kriging to behave like any other deterministic method
is important. First it shows thaL kriging can be modified to behave like standard
estimation techniques. so if they are desired, kriging does no worse. That is, krig-
ing subsumies the deterministic methods I investigated. Also, kriging provides the
capability to dynamically change the estimation technique. This capability could be
used interactively to adapt ively refine the estimation for different resolutions of data
and/or for different viewpoints of the rendition.
Kriging is considered the optimal estimator and since accuracy is important in
3D medical imaging, exploring the use of kriging to estimate values is worthwhile. Its
use has mainl\ been to est imate values within environments such as mining, gas and
oil exploration and other geo-science disciplines. Further research is critical to prove
the usefulness of kriging in 3D medical imaging. This research is also applicable to
any 3D imaging methods that perform estimation.

(i-.
Appendix A. Vanilla Marching Oubes Data Flow Diagrams and

Program Description

A. 1 Data Flow Diagramis


A.2 Progrom Descriplhon La nguage State,vi? (s

A-1
___~~ ~- a 42~~-z-
ORr --

F--

WiLLJ

0 TD

~~6)
C.D CD

0 m >

< 0

- oo

0
C/)
or -o CiO

KJ _

(0
LL
LiLO
CO
Cco
CE CO

uFu,.~
7L iI ,

10 LL

U))

-C-

CY E
* -o-

Q)0 0
-o (I
MwC
L1~ L(J)
P-V

CG
.G .sajis ssoid.. saqn,) 2uwip vl!"I wC- 51'
am'ui

0 (D
C7)

C0I

UO co
F CL

COC
QU) 0)
oo cc

0m C7
- -" D

C13

0 _ 0
0
0 _

C: c

0- / 0
010
0=_______
'Cl) Co

0 CF

L~ -1,lJ
06 -CD

F- CLl1
0 - C3 0) CLI~

IF-C~'-(0 -E Z

~~Cf)

cc
00
c-3

00

-00

0 ' 0) CIDJ

CD(D

Clo)
Primarv Data Structures

*plane[4] - Pointers to four arrays holding original p)oints and scalar values.
Need four in inemoiv t~o calculate normials using gradient operator.
"* ixPlane[2] - Pointers to two arrays 1hold1ing pre-interpolatecl Y points for cells
lbetween two slices.
* iylIane[2) - Pointers to two arra\,,s holding lprt-interpolated 'y points for cells
b~etween two slices.
* *izPlane - Pointer to an array holding the Ipre-iflterpolated "z' points for cells
between two slices.
" >'Gplane[2] - Pointers to two arra\ s holding calculated normals at cell vertices.
" gxPlane21 - Pointers to two arrays holding lpre-interlatd 'x' normals for
cells between two sli'ces.
" 'gyPlane[21 - Pointers to two arrays holding pre-interpolatedl ) normals for
cells between two slices.
* . gZPl ane - Pointer to anl array holding pre-interpolated Y normals for cells
between two slices.
* 256 case table.
* interpolatedl points translat ion ible. Indlexed by cell edge vertice.
* interpolated normals t rainslation labde. Indexed by cell edge vertices.
*isovalue array.

pfir4 comn
Clflfllim (*'7/) !.(rd wiuc
DE-D. ntinc ubbic 1)

while no more Commacnd lite arrineIit )j


-11 1 lilied by uiser
switch (Coil) Ia Ild){
case : sovalnilAg
reVad isovaljie argmiln(ut:

ca, : shelL-flag
reVad file nlaile aro unientl
readl isovalties fronm file. Einter isovalues Into all
array of isovaliies:

IA
case :widow-flag
read range of isovalues:
set as first. and secondl elemnets in arrayv of iso'-alues:

case : box-ilag{
set box flag to true:
I
case : Iist of..geom -iles -lag{
set list of geo-m files flag to I rute:

case : jpath-flag{
read path argumient.:
set path variable to argumett overriding default path:

case: data-file-fiag{
read from control file to get large dlat~a file information
such as imiage dimiensions. interslice amount.Input data, path. and
number of slices.
/'else default Is to read fromn au art ificial volume data file:-/

~ ~swit cli
}/* While

na d ht'qdt I-iynfo (" Tolp LI. re' OFD. Ihubbi, 2J

check (lata-filelflag andl open ap~propriat e file:


retrieve x andlv
if data-file-flag Iindicates. au art ifi'rial volumve, retrieve z (linensio0l1:
else if large (late file, retrieve ntiumbei of slices from coilt rol file.
Number of slices becomues ihe Z. dimiofn)'
N~
,4.2.3)
allocatf niemory slorage ("Top Level" DFD, Babble 3)

Allocate memory b~ased on dimensions;


/ * Example memory usage for arrays holding original and
iterm-ediate slice data.
p~lane arrays consist of structures with 4 floats.
all other arrays besides the isovalues array has 3 floats per structure.
Assume 256X256 medical images as input.
-1plane arrays x image size x 4 floats = 4 x 256 x 256 x 16 0141B
12 normal and interpolated arrays x image size x ~3floats=
12 x 256 x 256 x 12 = 9 MB *

)proceCs slices ("Top Level" DFD, Bubble 4)

A. .. 4. 1
nr'id slicts ("process slices" DFD, Bubble 4.1)

Initially read in 3 slice-, into ])lane[0].plane([1 1,l)lanie(2]:


D~lone t~o process border case, marching betweeni llane!0I and lplaned11. *
Then in main for loop, readl fourth slice. march

A4.2.4.2
.q / dclla diqlaiiccs ('"process slicc<" DFD. Bubbh 4.2)

/(]one after reading i first, three slices. befoi e maini for loop7
__\ (Is(plafl([01- > -x - (pla.ic([0 + I )- > xr):
-\) oli.(plan [0]- -> yI - (7plo2? [0] +x.dIi)?.%i0I)- > y):

AZ Eab(pl(~i7c[O]- > Z - planlfl->]- >


/note. this assumnes the delta distances will not chiange betweeni slices
and within each slice, i.e.. assumes a regular dlata grid.

A-8
A . 24.3
calculate normals at cell v'ertices ("proccss slices" DFD. Blubbl( 4.3)

Initially calculate for cells between plane[O] and plane[11.-


In main for loop calculate for cells between plane[1] and lplane[21:
InI final case. calculate for cells between plane[2] and plane[3]:I

A. 2.4.4
pr(-iflterpolalc points & normials ("procrss slices" DFD, Bubble 44
Initially calculate for cells between plane[OI anld lplane41I:
In main for loop calculate for cells between lplane[l] and lplaneI2]:
In final case, calculate for cells between plane[2] and plane[31:
Method :Use same comparison as in marching cubes for loop -

If a points' scalar value is greater than the isovalue, it~s a 1-voxel;


if one voxel is classified as a 1-voxel and another adjacent
voxel of the cube is a 0-voxel, then interpolate along the
edge and store in the appropriate array (which array depends onl edge);

A .20.4.5
uipdate translation lookutp table (-proc.-.s slices", DED. B7ubblf 4.5)

Prior to marching cubes (cells) bet ween two data slices. re-est ablishl
p~oinlters in tianslation lookup tables;-
/* These pointers must be re-establishied because the original dlata
slice lpoin~ters and the normal slice p~ointers are swappedl after every
new slike Is readl in from thle data file. Thel translation tables are
indexed bY thle cell edges and thle e'ntries cont ain p)o int rs to thle first
cell' dg.es. WVhen thle tables are uised, all offset froml the bt-4mbii ii g of
the arraYS Is added to tilie pointersN to access thle correcto cvL
For examiple tl~v[5][7] =izPlaii + x-dimcn.5ion + I establishes tile
absolute cell addr-ess for the Interp~olated point ol'cell edge 5-7.
A.2.5
march between slicts ("process slices" DFD. Bvbble 4.6)

/* input are pointers to two arrays of consecutive slices (planes). *


/ * main marching cuibes loop ~
.for(?*? = 0; iy < y dim:;?y±+
+)
.for(ix = ;ix < xdim7;ixr + +){
establish offset from plane pointers to lowver left vertex of
current cell;
establish all other cell vertices from lower left vertex;
calculate index for cell ("march D171", Bubble 4.6.1):
determnii i](jue case (*'march DFD", Bubble 4.6.2);
switch (unique index) {
for each of the 15 uniquie cases do
otjtut stats ("mnarch DFD-", Bubble 4.6.3);
get list of vertices from pre-calculated
table ("march DFD. Buibble 4.6.4)
get interpolated points k normals ("march DFD".
13tbble 4.6.:5)
otatit point,, k normals ("march DFD".
Bubble 4.6.6)
} e~
nd Switch *
}/ endl for ix7

calcudli midr.r for r(1l ("Iinarch I)FD. Bitbblc- 4.6.1)

test each Cell Vertex to see if it's scalar valuec is greater than tile
isovalue(s)
If a vertex is so classified, lboolvan OR a flag withi that, vertex #
Resuilt of flag aftepr all vertices are cla sified is the index into the
2.56 case tab~le.

A-1I0
A .2.5.2
deterimine unique case ("rn arch- DFD. Bubble 4.6.2)~

Access 256 case table with cell index, retrieve unique case #

A.2.5. 3
output stats ("march DFD., Bubblc 4.6.3)

Tally unique case #, ambiguous case #. of triangles and # of p~oints:

. 2.5.4
get list of vertices from pref-calculated table ("march" DFD, Buibble 4.6.4)

Access 256 case table with cell index to retrieve list of vertices;
Set a temporary pointer t~o vertex list array;

A. 2.5.5
gel intlerpolated points 8) nor-mals ("march"' DFD, Bubble 4.6.5)

/* For each unique case. the cell edges for interpolation were identified
and app~ropriate tianiflationq were selected, For examnple. unique case
3 requires interp~olationls along edges 0-4. 0-2, 1-3, and 1-5.
Two possible triangulationis can be chosen, but would not alter the image.
so the choice is arbit rar.\ in this, case. Now that the cell edges to
interp~olate along are knowni. t lie translation lookup tables can be accessed
wvith the offset calculated ab~ove. */
Access translation lookup table to retrieve pre-interpolated points arid
normals,11 along cell edges.

output points~ C vormals ("miarch- DFD. Bubble 4.6.6)

Out put points anl(l normals to a temporary points file;


Outplut line numiibers specifying points for triangles. Place in temporary
triangles file:
/temporary files used b~ecause AFIT geometry files requires all thle
points be listed first.. followed bY the point line numbers for triangles.
A.". 6
write box (-Top LevfsE DFD,. Bubble 5)

/ * only applicable to artificial volumies and only visiIble w~hen viewing


triangle mesh ~
If boxflag entered by user. write rectangles forming a.
box surrounding triangle miesh to temporary files:

A .2.7
output geometry fil (**Top Lce'ci DF[). Bubble 6)

Catenate the two temporary files, with the appropriate header and
trailer information attached:

A4.3 Data Dictionary Fo7- Data Flow Diagrams


Data Dictionary Symbology:
-is comnposed of
+ and
) optional
{} iteration, f)J+ indicates 1 or more. default. zero or more.
* * comnment

2-56 CASE TAB3LE = indices + {list of verticeq}+


* the pre-calculated tab~le of cell verti(es for the 256 different
cell cases. Each of the 256 cases contains a mapping
to a unique indcex and a list, of' vert ices corresponduuo
to the uinique inde. *

3D S(CALA R FIELD) DATA =*art ifical volumues, medlical imiage fides.


or scientific Volumue d]at a.*

l)ox = fineb drawn around dimensions of' image generated from)


artificial v'olurne.*

commands = (isovalue-flag) + (shelL-flag) + (window-flag) + (box-flag)


+(list-ofgeonililes-flag) + (path-flag) + (data-file-fiag)
*user mu11st '3up)ply onle of' isovalue-flag. dimell-flag or window-flag
isovalueilag - argu ueuut, is a single isoval u e.

A~l
shelliflag - argument. is a file name. File contains list of iso-valuies.
xviI)dow-flag - argument is a range of isovalues. e.g.. -w :30.0-40.0
lbox-flag indicates a box wvill be drawn around imiage - only used] for
artificial volumes such as embedded math functions.*
list-ofgeom..files-flag specifies that multiple geomn files wvill
lbe generated.
patl-fiag - specifies path for output geomietry files. Default is the
directory path of the source code.
data-ie-flag - specifies type of inpuit file(s) - 3D) SCALAR FIELD DATA.
Can be medical or artificial volumes. Artificial volume dlata, is default
and reqjuires re-directing stdin using i. For exampiille

(lelta..distances = * The lengths of cell edges. *

GEOMETRY FILE = AFIT geomnetry file of points triangle specifications.

header..info = data-imensions + number-ofslices + (inter-slice-thick-ness) + (in-


tra-slice-thickness)

indices = mainindices + tiniiue-indices, * indices into lpre-calculatedl 2.56 CASE


TABLE. *

interp~olated points k, norinals =*poini: and normals interpolated along cell edges
from- cell vertices to the isovaluie(s) providled by the uiser.*

list of vertices = * A list of' vertices is pi e-calilat ed manlually for each


case., in the 256i CASE TABILE. andl stored in the tab~le. Each list contains
the eight cell vertices, ordered according to uiniquie case ordering (following
he app~rop~riate conlplememtation and/or rotation(s)).*

points and normals = i poi nt,~ - [lie originmalI poi iii olfl Iedl froin the 3D)
'

SC ALAR FIELD DATA. normnals - Normnals ap~proximlated at Cell vert i es us"ing


(enit ral difference gradient otnerat ors.

slices =point,, + scalar-alm-s

slats =tally-anibigtioms-cases + t ally-inain-ca:4es -{- Ially-uniique-cases +

triangles = points andl normals


'fLU = Trainslation Lookup Table
* Contains pointers 1.o interpolated points and normals.
* see update translation table Jookup PDL for more info.*

TMP DATA =tip-points + tmpnlormnals + tmnp-rianglespecs

A\- II
Appendix B. Va'nilla Marching Cubes

This appendix describes the imlementation of the vanilla miarching cubes al-

gorithm developed by Lorensen and Cline in 1987 (34). It is termxed val.] !a because

my imp~lementation does not include any enhancemenits such a-5 texture' 7-apping

geometric ,olid modelling, or dlisambiguation. The first part of this appendix in-

troduces thle topic. Following thle introduction.,somne background inforMatinit ahunlt

decisions and about my implemientation are discussed. Then. the mnain step~s in the

implementation are listed. Next some background information on my iinnperenw-

tation is (discussed. Following this, use of the 256 element transformation table i.s

explained. After this, the cell edge Interpolation wethod is p~resented. Then the

method to determine triangle normals is discussed. Then. modificati)ons and en-

taincerneis niadle to thle public dlomain marching cubes codle Is present ed. Laslly.

(ow wlent s onl thle 1rmat of' the miarchinig cubes, on t~plt. are discussed.

1B. I1 Jlod(bChll

Ile Ilritail ri cuibe. algori tlrni (34 ) is ai 3D) inagilig nwi(t hod that exltract s at

Sutrface of i oterest hrorn a 31l) wolunle of (data. Th'le surface Is re[)resented a,; it :31)

t rianguIa r mnesh and rendered by a st andard polYgonal based gra phics reniderer. The

allgoii thiml lpocesse-, ci bes or cornpiu tat ional cells, w-here at c(ll is coinposedI of eighit

VON(+,. forur each froin two adljacent dalta SliceS. Each Cell IS aria Ivid to (let ermit~e

If' (hle surflace ol, Interest Intersects thle Cell. Triangfles are' geneorated wit hut cells
that. are found to0 containl a portionl of thle Sui a1C'e. Surface detection vithll a cell

is performed by simple thresholding. If a cell vertex is greater than the thresholO

value (iso-valure), it is assignred a one (considered a I-vertex), else it is assigned a

zero (0-vertex). Vertex classification in this marner yields 256 possibly different

cell classifcations. Lorenlsen and Clinie reduced this to 1.5 unique cell cases. Tri-

angle vertices are determined by linearly interpolating the voxel 3D points to the

isovalue between 1- and 0-vertices. 'Normals an similarly calculated by interpolatingf

previously derivedl cell vertex normiaIs.

B. 2 Background

In this research, I impllemented version of the marching Cubes algor-ithm

before deciding on a thesis topic. I thenr explored thle u,,e of kriging to somlehow

th suiac
inirov exr(tioi.Ii wadI \ilhelnis and Celder's (50) x\ork oii mnt a-cell

Scala r value est lillat ionl. Froml there it wa.- :3iiflfle to see Ihow kriging. colid be' used.

But. I still investigated ot her surface extractors to determinie thle history of marcingi

cubes all(I to see whetlici kiignig (-oi11(1 be app1liedl in ot her ai as. This led ime in1o

he a rea of :31) med icai imiagi ng, Itouw where mlarch ing cub~es as w~l as i nanv other

31) imIaging a Igordit iis were derived.

As sen fi oi chapt ci two, the two prima iy surface met hods ar-e en ben I Ic based

anld cell i niterj)olat ion. Tlme' ttieri lle- based approa lies were developed I~ Ileninam

Lill and Vd'upa in lte late 1970's and early 1980*s. In contrast to thle ctiberli le- based
methods. the -ell interpolation metiods, at-e inore li'ei'isticallY based. TIwo of thle

celi interpolation algorithms are Lorenscrn and Clines' marching cubes (34) and (8)

and Wyvill and McPheetcrs soft objec, algorithmns (.52). 'The latter two methods

are called cell interpolation techniqjues because they interpolate polygonal vertices

to the isosurface boundary along cell edges, \%here at cell is a paiallelepiped with

eight adjacent voxels as vertices. Cells are also known as computational cells. The

implort ant distinict ion bet weein voxel- based surfiace ext ractors such as cuberilI I based

miodels and cell-based surface extractors is the fo iner assume a cons5tant scalar value

throughout, the voluine lenevt (thle v-oxel). 'vherea, th? latter assumne a varying

scala r value throughout cells.

After reviewing the 3f) imiaging literature. I chose to continue eN1)!0riIn,1 the

cell interp~olat ion~ methlods for thle following reasons. First. I did not wvisht to develop

a specialized m, .. which would liit its it!,e o, f't .tre re;e'arch 10t potentially oiil

one a pplication. Cell interpolation Imd ho~is have beeni widely used in both mii-

cal iniagng and scientific visualizadviai. Add ;&l( all, the graph-t h"arecal inol2 ods

toed in thle cniheidle icnodlel are \ ery cuoiilx to ini)) lei and oftIeni yied inu.

that are jag'ged in alplparanne (Ibecaus~e of ithe 21) smrf ace display uitouedl -, uo

face) wit hout special shading pi-ocdre,IIcs 11C (1). 111 contrast. thle cell i nterlola-

ion methods(1 are much si mjlei lo uinder-st and an;d impjlemient an1(- all yield very

high quia 1itv in ages wit hoitt special sha Iillti ethlods nsvd . b~eyond(l he Ii adlit ioia I

app: (aclie such as (;oiiard or M~ono


g 13). Nost iimiport atly. since est ima~t ion is, a

BI3
critical part of 3d visualization, the ability to estimate %-,tryingscalar' values within

cell and visualize direct results of the estimation process 1)rie th1metst

investigate Improved estimation methods. The wvork p~revioully done by Wilhelms

and Gelder (50) provides a strong framework within which to visualize new estima-

tion methods. The work done here should have p~otential benefit to many other 3d

visualization methods. since piocessing cell-lby-cell is common to other methods such

asdirect volumne rendering (46i) and (51 )

The main drawback of the cell interpolation a pproa ches is the simplistic method

of segmenting the ob~ject of interest from the remainder of the volume. This method,

termed thresholding. makes a binary dlecision at each voxel - does the voxel con-

tribute to the final image or not? Volume methods do not make such a simple

(lecision, but rather, allow v-oxels to contribute percentages of different character-

inst ics (su1ch as color. andl density) to the fitnal imnage (33) and (16). Again. the

emplhasis of this research is not to compare volume and~ surface Intel hods. b)ut to

--onipare estimation techniques app~licale to bothI methlods.

beganl mv work with a public d('lmaini velnsion of a basic na rchiiig cubes

algorlihm. This version onlY set 11p t he pm c'calculat ed Ialle and performed thle Inl-

terpolat ioun t p. but1 (lid not dlo t hie inlio.s dithi( 1ill task. 111,11 of conipunt M.i~ iorninals

fromt gradlient. information. It basically d idsteps, 2-5 Ini 1he list presente(lIn time nexNt

sect ion. but as noted lat em. I re-wrole m


tost of the code to make it110 Cz unmdert andl-

able aind mnodifiab~le. As stated in dna pier 2. the imiaichi hg cube algori ihm -mnarches"
COMlRutational cells b~etween two slices of data. See figure 11.1 for a pictorial rep~re-

sentation of thle m~arching. The slices are labelled according to the order of arrays I

maintain in the implementation.

B.3 The Alain Sthp in the Implementation

Lorenson and Cline (.3-1) described the marching cubes algorithm in 1987. The

significance of their algorithm was that It used 3D information to construct Intcrslice

polygons to represent the iso-surface and to approximate normals for shading. The

following is a list tak-en from the 1987 article, which denotes the steps performed in

the marching cub~es algorithim:

1.- Read four slcsinto memory.


. Scati two slices and create a cube from four neighbors on one slice and
iou11 n('ighl iors o1n thle next slice.
:3. ( alcullate an index for t he clbe by, comparing the eight density %-aluies
at the ctiile vertices with the surface constant.
41. V.11c the index, look up the list of edges fromn a precalculated table.
5. U~sinig the densities at each edge vertex. find the surface-edge intersec-
ionl via linlear inlt erpolatio01.
0. C alculate, a 111it normal at each cube~ vertex imsinmg cent Ial differences.
Interpolate Ihe normial to each t riangle vertex.
7. Otpt flie t rianigle vertices and ve(rtex mormals.

Thie reiuainlder of tIls sect ion dliscusses hlow Sonie of t he above Step)s were uuii-

1 lCImi('ed. I do not (lisciss the steps that are st raigut fot ward from an understanding

of ilie basic algorlit, (esrilwd In chapter two of he t,-Ihesis-.


dfm

115
slice 3

slice 2

slice1 -7/
slice 0

j V~vxe i~jj+I x
~ i+I~+~
~voxel

V0lre 13.. C(omipuit atijonal (.(,] (Cube) marching ! ci weve dat a ;IIC(IS
First. I discuss the implementation of step -1in the marching cubes algorithm.

B.4 Pfrcalculated Tabil

Creating the pre-calculted table mentioned in step 4 is actually the first step

that must be done. This table contains 256 entries - one for each of the possible cell

vertex classifications. Each entry also has associated with it a list of vertices that

map to a unique case number (that also accompanies each entry).

To create the table. I first analyzed all the 256 possible cell vertex classification

cases to determine the 15 unique cases and mapped the remainders to the unique

cases (figures 2.6. 2.7, and 2.8 depict the unique cases). To simplify this process and

obtain accuracy. I reduced human error as much as possible. I used tinker toys to

represent a cell. with labels attached to the corn-rs and marked to indicate 'ertex

numlbers. I also use(l a presentation graphics package to out put 256 pictures of a

numbered cell wit h1a numbered segmented rectangle below to hold thelbinary value

of the case (See figure B.2 (a)). I analyzed each case by marking the appropriate

numl)ered labels for the case. coml)lement iMg the vertices if necessary. and rotaing

le cell to colre,plond to the clas.sification of a unique case. l'ach entrv in the 256

elemenlt table contains the oider of the cell verlices and the corresponding unique

case, The order of verlices corresponds the order of those specified in figure !B.2

(a) (0 I 2 :3-1 5 6 7). The ordering is arhitrarv but requires consistency.

I7
6 7 IF

2 Transform

0 (a) 1

1 0 1 1- 11 1 LRotate

7 65 43 21 02
Case 94 (5E hex)

Unliquie citse 25 (hex 19)


Vertex order 5 1 4 0 7 3 6 2

Figure 13.2. Examiple of cell mapping t~o a unique case


Aij example of how the table is used will help in understanding itb use. Assume

we have just narched to the next cell and case 94 (5E hex, 01011110 binary) is

encountered. The original orientation is seen in figure B.2 (a). In a case such

as this, Lorensen and Cline point out that "'Complementary cases. where vertices

greater than the surface value are interchanged with those less than the value, are

equivalant" (34:165). Therefore, figure B.2 (a) is transformed to figure B.2 (b). which

when rotated, matches with the same vertex classification as unique case number 25

(see figure B.2 (c)). The major table entry for case 94 contains 25 for the unique

case index and also contains the cell vertex ordering 5 1 4 0 7 3 6 2. Another table

called the translation lookup table is ubed to map case 25 intt, triangles base(] upon

previously ,.omputed interpolation points along the cell edges, and uses the ordering

in the table in place of the normal ordering 0 1 2 3 -15 6 7 (which is only used in

unique cases).

Step 5 is another place where )re-processing before marching can occur 1o

speed up the algorithm.

B.5 (0/1 Edge Inl(rpola ion

Cell edge interpolation is the process that determines where the iso-surface

intersects a cell edge. The cell edge must have one vowel valte gi('at er than the iso-

surface and the other less than the iso-surface for cell edge interpolation to occur

13-.9
One area where mnarching cubes implerentat ions call differ is whenl cell edge

interpolation occurs (step 5). My code interpolates in a pre-marching step. That. is,

every time a new scan plane (slice) of data Is read in, it is Immediately processed

to findl thle interpolation poinlts along cell edges where the surface is estimated to

cross. This requires a total of seven arrays, each with the dlimensions of a slice -

twNo for thle or''iial slice data, four for thle x and y interpolation values for each of

thle two p~lanes forming tile cells, and( one array for the z interp~olation values. Two

xand y Interpolation arrays correspond to each of the two slices of data that cells

will "march" between. Only one array is needed for interpIolationi points in the z

dimension, because of the geometrN of' tile slices anl the orientation chosen. Ths

c-anl e wen in figure B..3. which depicts thle correspondenrce between interpolated

p~oints along cell edges andl these array-s. The x27 represent points interpolated ,in

the \ dlirect ion where. i corresp)onds to the I th posit ioni in tile array, jis the slice

niumber. either I or 2 (thus 2 \ iliterpolatioil arrays). Not ice thle p~oints Only ie

along tlv' cell edges In the x direction. The, !1, points represent the same for the

y interpolat ion atiravys. Thie z,*s represent initerpolatedI poinlts ill tile z direction. If

tlie iso-surface does not intlersect a part icu lat edgIe. that coriesponii~g entry in the

array w~ill niever be accessed.

The following is the nt erpolatioll lo0ri1liul 1usd to fill t111he Intersection p)oint

along a cell edge. This formula applies oi] along a single component (xxy, or z)

sIince ai cell's edge lies only in one( of dth ire coordin ate directill~s.

H-if
uzcomnput ational

2" Y'cell edge

-101 21

x tilie ill positionl ill tile


de(1101e('
IJ - jdi x interpolation array, where
j = 1 or 2 and( i ranges from 0 to
(xdiml * Ydiml) -I

Fll
.tj p)ositionI ill Jtlh % inlerp arra%

i - it 01 positol ~ll
liSingle z iI1terj) ari'a%

Figure 13.3. ( orresl)0Inle of inierpolat ion arraYs to compul atioiial cells mat ch-
ino, bet ween~ Iwo slices
Given:

p1, p2 - the component points to interpolate lbetween.

vlIv2 - the scalar' value at points p1 and p2 respectively.

iso - the target value interpolatedI to.

i terppoinl/ = ((?iso - v2) * (p1 - p2)) /(vl1 - t,2) + p2

The translation lookup table is later used during marching to access interpola-

tion p)oints in these arrays. The advantage of this lpre-interpolation method is that

shared triangle vertices are guaranteed to be the same because only one edge i., ever

processed, which reduces computation time; wvhereas, during marching each internal

edge is processed twice. H~owever, the disadvantage is the memory required to mali-

lain arrays that contain the interp~olated points. Even more overhead is required fi

t his method to Iinplenment step~s 1 andl 6. which calculate Uhe triangle vertex normals.

B. 6 Nornial Calcuilation,,

My codle produces the normials during the pre-interpolat ion, step. TIe following

central difference gradivint operator is used t.o est itnate the out ward dIirection of' the

surface at a part ici lai voxel (I.j .k) along filie three coordinate axes (3,1:165)

13-12
Gj, k) =(D (i + 1,J. k) - D (i - L. j, k))/ A.x

Gij, k) =(D(i,j +~1. k) - D(i.j - 1. k))/ Ay

G (i,j. k) =(D(ilj k + 1) - D(i.j k - 1)/A

D(i,j,k) is the density, value at voxel ij.k and -Ax. .Ay. Az are the length!, of

the. cell edges in the corresponding component. Once the cell normals are estimated,

they are interpolated along cell edges to the iso-value using the same interpolation

formula used to find the triangle vertices. InI the vanilla marching cubes algorithm I

hancllc boundary cases as special cases when determ-ining the vertex nornials. That

is. voxels along the border of the enclosing rectangular volumre are assigned the

outward facinjg normal along the enclosing volume (see figure 1B.4).

B.7 Fixres to Public Domain Codf

The original code I started with1 was writ ten to be fast ( t houghi it was Jprac-

I icallv unusable because it did not produce wirnals nor planair poly-ois) withlout

regard to iriaintainabilhtyo i'nder-Standabilit


u v. Therefore. I had to re-enigineer a sig-

IIi -iauit,p~ortion of it, to obt ai nt


l 'oftwaic eii gieeriiig" goals. The il problemi

1he code had was that it (lid not linaint ai ii a t a ulardl logical order of lie origi iial

Slice., whli reading" t heml in from thle (la files. Only Ih luiterpolaw d planes were

Swapped to maintain orleI . Planles InI thle (code are (hle Salle as ai-ray correspond ig

to Slices. Planes are swapped to re-use the previously read slice for thle next process-

ilig loop of cells. Swa ppi iig lplawns is t he sMI1 asSWapping


Ie poi iters. BYswa ppliug

B~ l3
Fnclosing volurie

Boundjary case
Normal (10,00.0 0)

Bcuflcary case
0,maOOQC. -',n)

__ZZ
x

Figure B. I. Vanilla miarchinig cubes. ('ample boundary cases is urface normal

est
101 imal
all planes after each slice is processed. I cut the size of the translation table in half,

but it has to be re-Initialized after each slice ;s processed. However, the code is much

easier to understand and modify.

B.8 Alarching Cubes Output

The triangles are output in the form of an A FIT geometry file, which expects

all the points to be listed first. followed by a list of polygons whose vertices are

indicated by referencing line numbers of the above mentioned points. The AFIT

geometry file is then used as input to a slightly modified version of AFIT's GPR.

The code for the vanilla marching cubes implementation was written originally

inC' and consequently it was functionally oriented. It was later converted to C++

so it could interface with the C++. object oriented kriging code.

Appendix A contains data flow diagrains and program description language

st atements for the vanilla marching cubes imlpleeiental Ion.

I I.i
Appendix C. Cell Stbdi*vision Impienentahjon

This appendix dlescribes implementation (let-ails of the subdivision algorithm

inl more (let-ail than presented inl chapter four of the thesis. First discussed is som~e

terminology to help understand the rest of the appendhix. After that., the purpose

of cell subdivisioni is, presented. Then the cell subdivision Implementation steps are

dlescribedl.

0. 1 Terminology

Before proceeding. I present some terminology that eases the following expla-

nations. I consider the initial computational ct.!ls In the in~ain lou-- as major cells (see

figure C.1 ). The newly created cells wit hin a major cell. I term minor cells. Major

cells hax e the original voxels as vertices. so I con.lider minor cell vertices as ninlor-
IOxels. Finally. I consider the arravs within it a IoIU cela I)dvie
i-slices.

since logic'ally t hey i epresent. iilillt (data slices.

Q. Pi rupos'.' nf ( 11 .Su bdi:I. it

The p~rimllar1 1)111 (,r cell sub-d(i visionl ale 1o()disalihigilat e ambiguious Cell

Ca~ses, Mid~ to dleriye at bett er alpprOxima1lioll of the ko,-. irface. Ani aligiiols cell

obccuirs wheilever mlore than one top)ology can he Chowni fot he Cell. C'ell snhbdivisionl

(lisalihigiiates an) amilioI~s cell b\ S111)ivi6ding 'Iit( minlor celk. EBvn Ihoiigh t he

.subd i vided amibigill Il ni or (ell is nlo hon~et a i igi ii 'u ( hwca use 11 Is 110 lot gvi detll

C I
minor cell

Subdivision factor =2 iii all d-.-enisions

* Original voxlpls (major cell vertices)

0 Minor voxels (inor cell vertices)

Figure CA, Subdivided Major Cell

with). cell subdivision does not, guarantee that minor cells will riot be ambigruouls.

\Vilhclmns and Gelder (-50) handle ambiguous minor cells by choosing one lopology

b~asedl onl facial averages.


0* discussed in chaptei two. Th'e( cell subdivision metho
11(-Cin

also b~e used to increase image fidelity. This is done bvf uig the surface within

1ii inor. cells versus thle itch larger major cells - t ha t is.dila resolution is increased

by- foriiiing1 minlor cells. Increfasing Ilet data resolt-tion to imipiove iiage fide-lity

is the salme idea behlind the diViditig Cli) igol itlill (8). VVVI
('t )Oilit plinmiiives

are oitut 1 the t i angle-, out put in this cel


in thle dhividling cubes method insivadl ofl

Su1bdi vision mnet hod' Subd)(ividing call also be u~sedl Io icre-ase resol lit loll Iln onily one(

Tea ppen dix ent it led 1)isamibigim Iion ll En11lince~d Si rfa ct Pt-Inrentlat (",lSilid~i -
b%11l)
sw lmj~emls an e'xamlpie Ilhal helps, rlarif% %%h\ cell iibdi islioii cant Iimlro, imag, (11ti1lhN
dimension. equivalent to creating cubic voxels in the cuberille dat a model approaches

by creating new logical slices between the original ones. rIhis method is uised Ill the

impllemenitation of slice interpolation discussed in chapter four.

C.3 ImPk nClItalion SIP

This section lists the cell subdivision steps I implemented and describes each

in turn. Tlhe step~s ar~e:

1. Read data into0 memory.

2. March major cells between slices.

:3. Subdivide major cells into minor cells.

*1. Est imate ininor-voxel values.

5~. A pplv\ marching cubes surface ext ract ion with l majol cells" to forml surlface.

.StJ) 1 Ri(Id .Slic

Step I is basically the same step as, inl thle vantillIa mat chimit cu be'. (vuic) unI-

l)ielivt(it al in described Ill tlie preVVIOUS al)l)tIldiN. Fou


(Ali ateneded
MV illa it lM

to ca Icilate t lie gradenmt . The central (lifFreiene grad icitl opet atot l i (; oNxl
61~

vaile' sill rouinding"' a miaj"or cell for the calculat ion (sefit' .. This step) (lfers

f oil Ill(. X111Wimiplettetitat ion i H thle handillinrg of Ihorder caises. Thle vin, code d a Is

witl)hborder case, (Su~ch as, the first and last dat a slices andt' edges of data lc I

Its- pior#-s, of -,adieuu' ratciilarion wvithinl min1or relk, I, .n tlai.r setil~l


illose

:
approximating the surface normal at these major cell vertice-s wilth a met hod ot her

thani the central difference gradient operator. This Is because the 64 surroulilng

voxel values do not exist in these cases. Instead of the gradient operator in these

cases, the normals are applroximnatedl by the vector normal to the surrounlding p~ar-

allelep~iped shaped volume (see figure B..

Since two of the estimation techniques (tricubic Interpolation and kriging) usedl

tin the coll sub~di vision process need 64 surrounding values at all times, I hlandledi the

border case.- differently in the cell subdivisioni impllemefltationi. First. I always assume

there are 64 surrounding v~oxel values. To (10 this. I Ignore the dlata onl tile edges

of the 2D data slices used as Input. This is not a problemn because ]in most medical

Image (lata slices several rows and columns of edge values (10 not contribute to the

meaningful portion of the data. Because of this assumption, I Insure the artificial

volumes I (reat e are cent ered vi thin tie volume, wvit h at least one arrav posit ion inl

the x and y dimensions as a bunffer zone. Since I cannot ignore the first aiil last (lat a

slices. I create two a slices t~o replace them. The valutes for these dunimvy
dlmlv(at

slices are( copied from thle , ire I hiey are iiiat ilig.

Step 2 Is tithe same miarchilng t hat occurs fint die vmr implement ation. ( omiit a-

tional cells ar", miaichied betweeti two data slices (see figuire .1.1 ). fin cell suil)(ivisiOii.

iliese cells are nlot polygoitized, bitt ale subd)(ivided. This subdivi.iori is dlescrib~ed it)

lie iVxt St('j).

C
Stfqp 3 - Cell subdivi.sion

The purpose of cell sUbdivision wa described ab~ove - here the implementation

is discussed.

I process the subdivided major cell by employing another modified version of

my 'inc implementation to "'march" withini each subdividedl major cell. Since I use

a vmic imp~lementation, data slices are assumfedl to be read into memory; therefore,

simulate reading data slices inlto mnemory. Sice I never use miore than four data

slices at a time. I use four arrays to hold the m-inii-slicevausndpit.Ith

x'mc impler;.entation, data slices already have values and points associated with

them when read into arrays in miemory. However, both values and points must be

calculated for inii-slice a rnavs. The points are calcullatedl from subdivision factors..

specified prior to execution. Three subdivision factors along each of the three major

axes are set (e N.f= 3. N~ = 5. f7 = 2 meanis subdivide the cell in) the x direction

into three parts. iii the \v direct ion lino five parts.etc.). Figure (C.1 depicts a major

cell subdivided int1o eight min11or cells where the ,ulbdivision factor is two inl each

W(J in ach dIirect ion reqyIies thle ralculation of


dlirection. A sudivisin ftactor of tw

01iC 111mor1-Ivo~ed polit at thle mindpolit of eachi mla jor cell edge. one nhinor01-vo~el jpoilit

inl I lie center of each imia ior cell fare. ai oI0I il(,imioi- voxeA poinmt in)t he very center of

he major cellI. Figurie (C.2 depicts a hmajor cell subdivided ito 15parts inl all tIlirev,

Ii rect ioiis. The ca Iculat ionl for di viding a major c'ell edge into -5 parts is

IPL. = laJor.i + (7Ij07 '.2 - fl1 j.01,rr/:-


07
1)2. = mao.I+ 2 * (1)c(jI'1 2 -lll(1j x )/

':3..:r = ma 'o7-,1 +r 3 * (7nJ12( - 771(ljO. 1- )

p4I.x = majuri + .1 * (major 2 .X2 - maUjor', V/5

The y and z components ar~e calculated In thle same manner.

ma~ijol j)I x.~1ia jor X

'IIgmie (C.2. ( (omp)wtitig stibi)(lon


501poin~ts. Sul)(lI1 \i,1 factor =:5 in all the
(lirIect ions.

Once a mntor - oxel point 1is dc ferined. sralai va te i as,1-p ned to ite p.int.
.St(J 4. Estimate scalai' values

In order for the modified vmnc implementation to pr~ocess mini-slices, scalar

values must be estimated at the minor cell vertices. I implemented three functions

1.o estimate values at, minor cell vertices. These are trilinear interpolation, tricubic

inteepolation and kriging. Trilnear interpolation is explained in a separate appendi.

Tricubic interpolation is exp~lained in chapter four. Kriging theory is presented inl

chapter three and the inmplermentationi details are in chapter four. Once thle minlor

cell vertices are estimated. the surface canl be formed.

Step 5. Apply marching ciibes quiface (xtraction vwithin major cells

I implemented two modified versions of the vmc inmplementation described in

the previous appenix to acconlhish cell subdivision. The first vmic implementation

is usedl in the outermost. l00o) to read1 the actual data slices inito memnory. rhere are

two pri rnai v loops in the systemn. Nkit Iiin this out eimost loop, imajor cells are formed.

The only tasks thec first, \ inc imnplemenitationi does is read dlata into memnory arid foirmi

major cells. The surface is act uallv formied by the second ;nic impllemnentationi. which

p~rocesses 111a.j0r cells.

Thie secoiia uc iminplemient at ion t teats major cells a,, stil)-\xol iiiies. ext ract intg

sub1-surfaces from them Thius the secondl primary loop nia rches," within major cells,

Aii ng p)0rt ions of the iso-surface. Tliat is. for each majoi cell wi ii a voutme of 31)

dat, the second vmic irnlplcnmet atioii ext ract s a suba-sui face fhoin each stinbvolumle

(major cell). The major task htis IS o ilislc I li o ita (oniti nots surface ext ract-ed
not only b~etween nitor cells but also between sul)-volumes. A continuous suirface

means the triangle vertices and normials are the same at shared locations. Fortu-

nately the vmc imp~lementation described in the previous appendix insures surface

continuity b~etweeni minor cells by a pre-niarching interpolation step. Hlowever. the

task of' insuring surface conitinuity betwveen sub-volumes is not as straightforward.

The challenge of dlealing with inter sub-volume surface continuity (leternhtines

how miinor ce'! vertex normials are approximated on the boundary of' major cells.

Recall fromi the %mc algortithm that surface normals are approximated at cell vertices

by a central differenlce gradient operator. I use this op~erator to approximate normals

at all minor cell vertices comp~letely contained within a major cell. However, the

inor cell vertices onl the borders of major cells are handled specially to insure inter

sub-volume surface continul-ity. First. those minor cell vertices that are the same as
the ma jor cell veirtices are assiglmed thle same normal value as the niajo c elli c'.
C

Tihis ISpos.sible because pie- iiiarching inteirpolat ion of both points and normials is1

perforined inI the outermnost loop. Then. prior to marching within a miajoi cell. I

calculate cell fa ( -norm a Iaverages and cell edge' normial averages to use oil tlie o1 hci

bou1 ndary cases. T[hese norlinal averages shared bet weenl su )- vol u ii es ilistiles ii ite('

v.
su rface suili \'ol[IIII(, coiit iniiit*1

After the points, val ties. and noria Isaile e!simat ed for- the in inioi cell \ elICC's.

l then inlter'l)olate pollit s anld niormals in a pre-niarch i g step (the sai ne as in lie vI](

inplelleit (Ition
III tlie previous cha p1 ei
decCri bed InII to deterniilie thle sur1face- ni nol
cell intersections, Next. I -mnarch" mninor cells between mini-slices and output tran-

gle vertices and normials to an Air Force Institute of TecThnology (AFITr) geometry-

file. To render the Surface. I call a mod~fiecl version of the ARIT General Purpose

Renderer (GPR) to do Pliong illumination and Phiong shading. Anot her applendlix

describes the modifications to the ARIT (;PR.

1 (C-9)
Appendix D. Disambiguation and Enhanced Suface Repesna 7io

by Cell Subdivision

This appendix describes how cell subd(i vision can dlisambigua te ambiguous cells

and how it can enhlance the surface representation. First. some b~ackground inforina-

tion is presented, then an examiple is lpresenitedl that helps demonstrate the p)urp~ose

of this appendix.

D.1I Background

1as well a-, \'NV.ilhelins and Geler (50) demonstrate that subdividing cells

redluce-, amb~iguity significantly. and depending on the estimnation function used to

estimate( i ntria-cell scalJar values. ( an cause a smoother represent at ion of thle surface

generated by cef llitci pola; ron. V sing a subhdi vision factor of 5 initach (Iinicnsion.

I explore b)ot t Wii er and triculbic esi mat ion funict ions ini at tifici al volumes Thie

tri linear function generates a 1bettIer surface fit than the vaiilla mtarchinug cu bes. but

is still far Fromt thle desired surface. 'Iricuit i(i et imna tionl event perfortm't bett er in these

anti ficialI voltunes. 'Ii icublic estI iIla Lion causes Ihle surface ext ract ion to getterrit a

Closet repi esent atiot of thle actnia surface I Itait he 1.Y i-fitear d(oes. IThe aut hors cited

above cl aimi the I ricu bic is,bett et at est'i ria Iinug )oints wit hini the' cell beca use it ises a

larger neighb~orhoo~d of points wit iout assutntg Iinva ri t H owever.~ this assumttpt ion

jtta~r rot be \-ali(l foi chIt a \\it It slim rp conrnast s withIini a smnal Iiteiglibot hoodI of \o\C5.

I)- I
A larger n~eighborhood may in fact cause erro~rs in data with sharp) contrasts. lKriging

estimation canl also use a larger neighborhood of control points, Thle promising nature

of kriging is that it guarantees the -best' linear estimator and the neighborhood size

canl be modified as well as a number of other lparameters.

Subdividing amlbiguous cells does not guarantee that amb~liguouis cells will be

removed. The minor cells created from thle subdivision p~rocess may lbe ambiguous.

\Vilhelnms andc Gelder (50) apply the facial averaging techniqjue, describ~ed in chapter

two. to disambiguate minor cells. That Is. i1'a face is ambiguou-. the average valuie

obtained by av-eragingr the four face vertices is tested against the iso-value. If the

average is greater than the iso-value, the 1-vertices are connected. else the 0-vertices

are connectedc.

In all the artificial dat a sets I cr( ate. ambiguity is completely removed with-

out the need for (lisambliguati uiorCells. llowe~er. tis does not occur. in thle
1111g1

Medical data sets I test-ed. In the mrtificial data set., the surface gCnerated by cell

initerpolat ion ap~pears smoother the higher the subdivision factors and depending

of) the e'st imat ion f[iiit ionl. "I'lenext sect ion eXjplores how is Smnoot her sulrftace

rep~resenitat ion canl occur.

1). 2 bEww p/

The Jpolygoniza tion in thle miarchin g cubles algorit hin or all\- cell interlpolatuon

algorith1mm is a rbitra ry. It IS I gule"S at how thev Surface should pass)t~


t hroughi thme

D-2
Figure D.1, Alternate p)olygonization for case 6

cell. For example. another triangulation of unique case 6 of' figute 2.7 is depicted in

figure D.1L

Note that only 1hW a nhbigUOUS CaSes of the uniiique case figures in cha pier two

wte 1ru I andhig iu in the sense t hat ahIt iate fpol~gonizal itns (,ani be pei forned

For exatlel)1. fte' polyon1iZati on of unique case 5 (a non-ambiguous cabe) is fairl

ob)\iout". eveil iough if is reniotely possibIle furt her suit)livision aniid st iinat ion coti 1(

unicover a dlifferenlt top)ology'

11) uttdri aid how cell sli)diiioti and emtitaial cani (liatil~gtiate ani ~in-

otis cells. considier a major ambiguous cell in which a pibil( erroiteots topology is

gerieated (mntajr (''i~ elr to an uninixivided t eli. a in inor- cell a sitbdcivided cell).

'Topolog r, fer- to0 the 1)olygoization of a cell t hat reprlmt sii a lsort lol of dic Sirface Iop0o

IY3
Without subdivision. erroneous top)ology is at gross error. which not oly~ (cfli'es ii-

accuracies Ini the final Image, but canl also create *-holes" as discussed Ii hapter two.

Sub-clixiding this cell will reduce this one large erroneous topology into smaller. cells

where most wvill have iioiiamlbiguous cells, dlepending onl how well (t( e estinliatiOli

function estimates the surface in the cell. Case 14 (see figure D).2) Is a particularly.

goodl example of a rare case, even if no ambiguity results. The reason wh) it is rare

is becaLISe it repi esents aier compl)icated portion of' the Surl'ace topolog.1

Suppose case 14 is subdivided by a factor of 2 Ii all three diiections. The

chances Of case,( 14 app)earing within an\y of the minor cells is even rarer. This

is so because of the complex triangulation of case 14. Assume the tiangulation1

of figure D.2 A depicts thle surface correctly within that, cell. Trhen subdiivliing

the cell could possib~ly generate the subdivided major cell depicted Ii figure 1).2

13. Ini tlk case the topology remnained the satme. and caise 14 dloes not show upl.

In l*a(t. all iioiieiiipty tumior cells are unique ioiiaiiibigulous calse I inl this figure.

Ini any.\ su i)divisioii. the 1-vertices of' the major cell wvill reiiiain 1-vertices iii t he

corresp~onin~ig i ior cells because their values (do not change". I lowevelt(-\\.ex I-

\eii idces may be added oii Ilie iinor cells. Iii th is figure, h I 1-ver-tices, are thle

samlle. F'on sinll) licitvy. I wxiii discuiss otik. one edpe of Il( iuajoi Cell witere, Ilie siirfa c

iiiterseci s. edge is(diot ed b Y E figuriDI).2.


lI.i

Figure D).2 C depicts possi blY different minor cell topologies caused bY lhe

suibdix-ision. Note in figure D).2 13 that the surfac itc enli


scct bet ween minor VOxels
0a

'If)

C intersection
point

Figuire D).2. Lx i )(~of altei ndl i NfA C rejptCsetitathio Cansli I, rd!I WANK%
ViiO
a and 1). Figure D).2 C however shows Iilore complicated 1minor1 cell vertex classifi-

cations. where the surface intersects between mninor voxels 1) and c. InI figure D).2

C. thle upper right front cell is now unique case 3 instead of' I and a niew%nonemipty

mninor cell exists - thle lower rihtont cell. This cell is atot her' unique1 Case I. Thi1s1

does not imp~ly that figure ID.2 A is a wrong topology, it just means that in this case.

figure D.2 C cap~tures the sirlace intersection point on edge E more accurately.

Of course there are miany other possib~le top~ologies wvithin thle iini1 cells, too

numerous to list here. The point of the simp)le example presented is to shiow that

sliId~v'son (-an (lisamlbiguate and] cause the cell interpolat ion to provide a closer

applroximat ion of the actual surface. If m-inor cell vertex value estimation is accurate.

the new surface itersect ion points should be closer to the true surfa-ce boundary. thus,

generatin a triangular mnesh that better applroxiimates thle actual sur-face of nil (W-est.

S1ubdividing t he celli even [fi rt bet sIholi d generat e evenl closet' su rface iitersect ionl

pint s. Againi. resoluition of atnibigitous inor cells Is not guiaranteedl. buttits stated

previouisly. t he few. ainbigitomis jinntor cells hlat may result1 canl be dealt wit It by fa( ial

aveagng g~iiettonsist efrv hieuIristis or sIIiilarI.et1 hods.

1) 6~
Appendix E. Bzrtary Image Format to Utah fiLE Format Convers'ri71

It is often very usefull to look at just. a single slice of data. especially if tile data

is from CT or MNRI scanning technologies. However. most of the data is inl binary,

so it must be converted to an imnage format. I chose the Utah RLE format hecause

Ican View anl RLFK, Image from any of time different types of workstations we have at

the Institute. Mv code a-ssumes binary files aN inplut. but canl be easily modified to

read from ASCII files. Tile code is in the directory tl.esis/src/nakerle

Thle majority of Image files are stored as I bytc per value. with values stored

inl scanimie order from bottom to top. 'Theref~ore the double for loops used to readl

andl write from/to thlese files increases fastest MiX.

Somle differences ill file form11ats were discovered dI(iing t hiis effort. The C'hapel

l11ll data lias 2 bytes,per valite. II li,


hiseS. it Is imiport ant to determine if the bytes

nleed swapp~ing because of architlecture dlifferenices (little endlian/big endlian). Also.

s"oIe MR]I inidge dat a derived frouii anl uikn)wii om-(


r as st oredt as stil iaster file".

TO (letermlle If I he Iimge format is 'uuIIi ra't ei I iy I lhe s!In call (onl a Sunl Coiisole)

.%C( Iflf/I IIIia'1V-f1ilV> Orl- I niN COIiiiiiatIId ]ih.

Thc I t a ii HI., 1Koolkit has !- -Itines, that ditIlie same task (gi'avtorle and raw-

ore): however. Ithe code I d- .-


lopedl allowed Ii(- to Jperforil 01 her operations onl tlie

(data. For example. I miodified if to) omiplite the histogim oIf tilie
(data. The miajor
benefit of the code is that it has the basic format for reading binary data, which is

necessary in other areas, such as the actual rendering piogram.

t'-
Appendix F. Changes to the Air Force Instte of Technoloqy

General Purpose Renderer

AlFIT's CPR is anl ob)ject orlei~ted renderingssystem written in C++. It has

several hidden surface removal. reflection model, and shading model i mplementa-

tions. This p~lethora of' options coii'ecjuentlv makes, the executable qjuite large. Since

mlemlory Is a slcrc reIsource whenl (lealing wvith volumes, I decided] t~o minimize GPR's

imemory- usage. I (lid this by removing unnecessary p~ort ions such as code handling

text ure mlapping, vertex colored polygons, Bezier patches, scanl line z-l)1ffering, etc.

I retained the X buffer and 'A' buffer hidden surface removal Imnplemlentat ions, all

the rilection niodels, and both flat and Phiong shadinig.

;\lthomgl this reduced the execut able size by over al half. C PR still required

too miuchi ililmorv for a single ge.olmet r\ file outpult froml mly minlungll cub~es im1-

lelien tat loll. I at Iemp? ed to fix this p~rolem~ by cineat iu a list of geoimet ry files.

sI( ; P l? is capa l le of reading mini It iple fi le". "Ihe code is sup~posedl to hev immemnlorv

a ft em proces,,Il,ugach geumuet ry file: hmowev-er. t he dea Ilocation (lid n ot a jpjeam to wolk

corrc
I.I((l v.

C IPR allocates many arrays. It Is possible thle GIMN ('+ + a Ira)\ (lea io(at lOll

doe., not work properly. W\hen array deallocation is at? ('ipt (' according to Schildt

(38 :337). t lie g+ + warmring mess'agre "amray size vX Jpre~si;lt fom (deete igniored" results
(e.g. delete [pcouiit] J)IltiflveCts). Schilcit indicates the importa;,ce of' t his operat ion

(;giiored by g±+):

One reason that youn ieedl to specify the number of elements in thle ar-
ray'N
to the delete operator is so that the proper number of' destructor
functions can lbe calledI (that is. one for each object in the array).

Hlowever, after I Inves~tigated memnory allocation an(l (leallocatioii by creating test

case., using thev same dlata st ructures usedI bY GPR. I discoveredI that the arra.\ size

had no affect onl leallocat ion. The main factor appears to be the order anl(l tilie

sizes of the memnory blocks allocated. I found that if a smaller block of imeinoi isl

allocated first and then freed. a larger block cannot use the just freedI space because

it is not large etioughi (both1 malloc( ) frce( ). and new.delet e appeared to operate the

samle). Th'lerefore. for thIe best use of U nix m'leniory1 i ia nageiniei it. large llock.shou l(I

be alIloca ted fi rst.

T]his kniowledge still did not solve ilhe p)roblem of GT


l ui
wH too10much nienlor\.

becaw
u GTCPR alIlocates ma nx block, of' \ diryi n, size. based] onl lie polygon count ti
he

vertex ('011111 aiIn ui1viimh11er of, yerti(e(' per polygon. To k~eep fiomn alt ci iiia( G'Pl{

sign ifica ikl to order mnillrv al loca tions (0 (c ~.I ch(Js(, to imiiplenment a \c rv

Simpille inemory iliatiiigenient schieme.

The aVerageV lMeMoi requirelld fori.oEsiiggoie yflsoi u froiii mnarch-

mug cilbs is bet \\vem thbree all( fike mnegabytes per file. If (IPl successfully miikes

le in l he'li spare aft cm e


fis aimnonntl ;i'~ail~il file is lproc('se(l. I lien no umevnorY
pro)blem~s result. Thierefore. I dlecidedl to e-iat my own heap by simply allocatinig

one b~lock of inerory large enough to handle the largest file in thle list of files, out put

fromt marching cubes. This size varies based on the isovalute(s) selected anl(] the itit

dlata resolution: howevei . I found the niaxinimn needed never exceeds 10 megabytes.

The memory block is allocated once at the beginning of GIPR it)n ain( I~by

nwirifpt i = (void ') malloc(MIOWY):

where MEMNORY Is the lpre-defined size.

Since memptr, is type) casted as a voidl '. portions of this block can be Cast to

an- t pe. An offset into t his memory block is miaintainedl for assigningy newI me]mor10.

B~efore each geometry file is jprocossed. this offset is re-set 10p zero. Thums. It0 (lelette

or free olperat ions are( nereh~aiX* anild te samev nivinorv is re-tised over aII(I over.
Appendix G. Creating A rtficiai Volumes

This appendix discusses the merithods that create an artificial volumle. Ani ar-

tificial volume (term taken fr ii Tiede (43)) in this wvork is one in which scalar

values are artificially e.,tered at node points in a 3D array to represent some object

surface, versus a volumie containing voxel values which are gencrated by a. scanning

technology such as NIP] or CT or by a scientific simulation. The primary artificial

volumes I created for thi.5 work contain surfaces depicting three-dimensional mnath-

emnatical functions such as a sphere,an ellipsoid. or a p~araboloidl. In both methods

I implemented, the surface is always centered within the first octant (positive x,.Z)

by subracting the center point (h.k, from the points 'x.-v.z) in the math equation.

The Initial method I clew loped is very straightforward, but does not allow

setting at surface threshold ot her Ihan 0. 1 accompJlishied this I)Y assigning to each

voxel the value returned by evaluating the mnath function at the voxel 3D point. For

examp~le, a voxel value at miesh point (x.y~z) for a sphere Is determined by:

V~u(x
VQC 4 i)= II( )(.7-
=" -h)' f (y -k) + 1-)' -r

where r Is the radi us anid (Ii.k .1) the center point,.

Funct ionis definling it Surlface retum'i posit k e values omn one side andl negativ.e

values o1] (lie other. wvhere stirface points are evaluated at, zero. Since nmarchming

cubes interp~olates triangle v-eilices to the sralar value bet ween voxels, this met hod

generates it volumiie In t he correct formnat for miarchinig cii bes to read . 'T i~-is so
l)Ccauso a Lifl'ace defined by a mlath functAin Will rare]\, if'ever Intersect ani art ificial

volume voxel. Of' course, the larger the volume, the b~etter chance,, this will O(cur.

Lt (ol Phil An.burn extended the inethod just dIiscussed to allow any scalar

value to represent the surface. Since the valuie returned from a math function evalu~-

ated at a niesh point denotes the distance from thle surface in a positive or negative

dlirectionl. this distance is used to taper off a value from the scalar value chosen. Also,

to get the out put in marching cubes formula. I modified the algorithm to taper off

towards the negative if the point is oin time negative side of the surf'ace and towards

thle p)ositive if tile mnesh point is Onl the p)osit ive side. The formula for this method

to (letormli n thle value at a mesh point is

di= fabs(val'ue~x, y. z)

if (c/ x,. )> 0.0)) /* positive ~

co~rdl
(I/Ut(a. 1/.:) I I,INDS
1)1.,, T A

+(1~IT (SC'ALAIR?+ 01-VS I DE)

ekse /~Imegat i\e

'oxrcllu((.-.y. z) = I )*.-I LAJ

I I.USed A HiTs viewit program onl thle Siliconj Graphics :3100 series workstations

Ito view time artIficial voluimmes. I also iliciided a conmmaund inIe Opt ionl t~o draw a b~o.\

arli~il t h11\()lulme p~eriimeter.

G-2
Appendix H., Trilinear Inteloaton

This met hod assumes scalar values vary linearly along complonenlt directionS

between \voxels (45:11-14). It Interpolates in each of the three dimensions, see fig-

u,-e 11.1, where 1'(01) and f(rii) represents the scalar value at original point i and new

(or intermediate) point i, respectively. The goal is to estimatec f(n6) given f(OO) -

f(07).

T he method complutes ratios of distances aind scalar' value dlifferences succes-

he
utilvaue
ente
sivey i deermied.Thatis PI-fOO) = f(O1)-flOO)

dll

The!I. f(n 1) IS found by Interpolating between f(02) andl f(O03) )inthe same

manner. f(nD) Is foundi~ 1.\ interpolating between the Intermediate values f() I ) andI(

Q(i:3). The f inal value 1(116i) Is determilned bY interjolati g bet ween) f'(114 and f*(u).)

There is oneC SIgnificant jlrOlblein With thiS met hod that redCeIICs accuiracy of

est im~a tes Iii a t hree-dimuensionial dlata set. Only the eight surirouind~ing voxels are

aim k Zed 'o estimate a new va ime at a part icu Ia poinit wit i in the cell. wh en ii fact

IhemeI
eit sam ples muay not provide suf ficient inzforniat.ioii to infe I lhe variahil it \ of'

he data. Nlore impoit ant ly however, thle lat'a may not1 vary lineairly along" major

cell edges.

A 1)101 dinI of' less signficance 'is the direct ion InI whlichi to in Iei'polat e in it ially

is .111 t a I a ri i)1ia rY assuI 1l)t1 ion (I Ilotough oly tY


I.II ee (1ilcc., ex Ist). T I s is IIIIport alItI

Il-I
because f(O-) arid 1'(15) are based on the first set of intermediate interlpolatio)s

in this assumed directioni (e.g.. the Y direction could be assumed initialY - thenl

f(nO) Would be determined by interpolating between points 01 and 05). This is a

potential cause of inaccuracy because Jor example, values may not vary the samie

letween voxels 00 and 01 and between 01 and 0.5.

f(02)

Fire 11.1. Triliiuear infer)oial iou

II2
Bibliography

1. Artzv, Ehud, et al. "The Theory, Design, Implementation and Evaluation of


a Three-Dimiensional Surface Detection Algorithm,": Compjuter Graphicq and
Jbnag( Processing. 15:1-24 (1981).
2. Artzy. Ehuci, Gideon Frieder and Gabor T. Herman. "The Theory, Design,
Implementation and Evaluation of a Three-Dimensionial Surface Detection Al-
gor1ithmi," Computer-GrapThic,;; 14(3):2-9 (July 1980).
3. Brodkin, Chris. The Application of Kriging for Controlled Aun imiization q f Lavge
Data Sets. Draft MNVSthesis, School of Engineering. Air Force Institute of' Tech-
nology (AU,), Wright- Pat terson AFB3 OH, December 1991.
4. Chen, L.. et al. "Surface Shading in the Cuberille Environnment,"' IEEE Comn-
p utvr Graphics anad Applications, 5(12) :33-43 (December 1983).
53.Christiansen, H. N'. and T, NV. Sederberg. "Conversion of Complex Contour Linle
Definitions into Polygonal Element Mosaics." Comnpitter Graphics, 12(3):187-
192 (1978).
6. Clark. Isobel. Pin ct?cal Gostatistic.;. London: Applied Science Publishiers Ltd,
1979.
7. Cline. HI.E. et al. -3D Reconstruction of the Brain from Magnetic Resonance lIm-
ages Using a Coiinectivit\ Algoithin." illaf/ndetw Rt.sonanc( Imagng. 5(.5)::34:5-
-352 (1987).
Ch(le. Harvey E.. VI lliain E. Lorewn etc al. '-Two Algorithms for the Three-
Dimensional Reconstruct ion of Toiogranis.- ;Wdical Physic-s. 1.5(3):320-327
(May/Junec 1988).
9. Cres'sie. Noel. "ErifyingNonst.ationary Dat a," Iow-val of I&e American .Stall.,;-
Itcal A.1.soctatio,,. 81 (395):625 -63 1 (.Sept embler 198(6).
10. ( ressie. Noel. "Spatial 1huedict ion anld Qid ilnary Nii g. 1althe at iral Ge ol-
(My. 20(l:405-121 (1988).
HI. (re."ie. Noel. -Geostat isti( s." 1h A iiiercan 'SIahtician-
.p(4 ): 197- 202
(November)C 1989).
12. (7ressie, Novl. 11wi Origins ol' l\riginig." A at/i(fiat;((Il C ology. 22(:3) :239 .59
(1!990).
1:3. lDavidl. Michael. (o%tat,.stwal Or( I?c.S( rtc Estimation. New York:Eeve
Scientific Publishing ('ompany. 1977.
It. C. .$thitI.i(.- and11Data Anii (1ISs 171 (iCOlO!/! (Second~ Edit
D~avis. .11)hn1 ion). Ncw
York: .Johni \Vc."Iev kY Soils. 1986~.

13111-
1T5. Delfiner, P. and J. P. Deihomme. '-Optimum Interpolation 1)\ Kriging." Display
and A nialysis of Spatial Data edited by John C. Davis and Michael .1McCillagh.
90-1 14. New York: Johin Nkiley &~Sonis, 1975.
16. Drebin, Robert A.. et at. "Volume Rendering,"' Comnputcer Graplic,;, 22(0)65-
74 (August, 1988).
17. Dubrule, Olivier. "~Comparing Splines and IKriging," Computers & Geosciflnces.
10(2-3):327-38 (1984).
18. Durst. Martin .J. "Letters: Additional Reference to Marching Cuibes.' Con?-
puter Graphics, 22(2):72-73 (April 1988).
19. Farrell, Edward J. and Rosario A. Zappulla. "Three-IDimrensional Data Visui-
alization and Biomedical Applications," CRC Critical Reviews in Biome~dical
Engineering, 16(4) :323-36.1 (1989).
20. Folev. .Jamies D., Andries van Damn Steven K. Feiner and .John F. Hughles.
Corn palter Graphics: Principles and Practice (second Edition). Addison-Wkesley
Publishing Company, 1990.
21. Fox. John. Linear Statistical Models and Related Metdhods W'ith Applications5 to
Social Research. JohnD Wiley & Sons, 1984.
22. Fiieder, Gideon. et al. "Back- to- front Display of Voxel-Based Objects," IEEE
Comnputer Graphics and Applications, J(I):52-60 (Januiary 198,51,
2:3. Fuchis, Henry, et al. "Optimal Surface Reconstruction from Planar Contours."~
Communications of the A C'M 20(10):693-702 (October 1977).
21. Ganapathy. S. and Ti. G. Dennehy. "'A New General Triangulation Method for
lllar ont olw sF. Comnputer Graphics. 16'(3):69 7-1 (.1 uly 1982).
25. Grant . Michael. The A4pplication of Kriginig in. Ili Statistical Ainalyspis of
.4ntbroponictric Data V1olaime I. MS thesis, AMIT/GOR/ ENY/ ENS /90M -
S. School of Enoineering. Air Fo rce Institute of Technology (AU). Wright-
Patterson AFB3 OH, May 1990 (AI)-A220 61:3).
26. Ilerna n. Gabor T. . et al. '"(omipiter Trech iiqles fol. tile Hepresent at ioi
of' Ilrev-diniensional D~at a on a 'Iwo-dim-ensional IDisplav7 SPIE'. 367:3 -11
(]()8-2).
27. Hermian. Gabor T. "A Survev of 31) Medical IniaginugTechnologies.- IFEl:l
PF*tginiing iii Mlicine (and Biolo!/. (I):15 17 (Decemnber 1990).
2$8. flernian, Gabor T. and( .Jayarain 1K. Uduhpa. 'Tisplay of 3-1) )iglial lin-
ages: Comnput at ional Foundations and Medical Applications." IEF,' ('om puir
Graphics and Applications, .39-46 (August. 1983).
29. .Joinhel. Andre ('. Findamcntals of (kostaistic., lit Fir( Lrsson.%. W'ashiington.
1). C.: A\inericall Geophysical Un ion. 1989.

1311-2
:30. heppel, E. "Approximiating Complex Surfaces~ by Triangui-lation of Contour
Lines,*' IBil Journal of Research and Dcrelopincut. 19(1):2-11I (.January 197.5).
:31 . Xerbs, Lynda. "GEO-Statistics- The Vairiograi." COGS (ontc , Cont ribu-
tions, 12(2):54--59 (August 1986).
:32. Laur, David and Pat, Hanr-ahan. "Hierarciiical Splatting: A Progressive Refine-
mewnt Algorithm for' Volume Rendering." Cominpthr Graphics. 25(4) :285-288
(Jidv 1991).
:33. Levoy, Marc. "Volume Rendering and Display of Surfaces from Volume Data."
IEEE Computer Giaphics and A4pplications, 29-37 (May 1988).
3-1. Lorensen. \'illiami E. and Harvey E. Cline. ".Marching C-ubes: A High Resolu-
tion 3D Surface Construction Algorithm." Coinputhr Graphics. 21(//):163-1 69
(July 1987).
:35. Matheron, G. "Principles of Geostat istics." Economic Gco/oqy. 58:12-16-66
(1963).
:36. MNcCormlick, B~ruce K1. et al. "isualiation in Scientific Coinpitingj Computer
Graphics. 21(6,):1-7, A-I to 0>18 (November 1987).
:37. Mendenhall, William. et al. A11laihemnatical Stati-qics ith Application,.. PWVS-
1KENT Publishing C!ompany, 1990.
:38. Sclildt, Hlerbert. C~++., th Comnpht. !?cfti nc. MchGraw-h1ill. 1991.
:W9. Stx-tz. M. RI. and 0. Frieder. "~Tlree-Dimensional M-edical linaging Modalit ies:
Anl Overview,"~ Critical I~~ 4in? Iioincrial Enqinu ring. I,1 Issit( 1:27-541
(27-54 199().
10 . stvtz. MailiiR. am I Oph~i lrieder. -CThupuit c Sysl eni fo 'I'llrev- Di ilucnsioialI
lDiagnost.ic Imiaging: Nnii aminiat i of thSate of the Art ." Crficti!a ThRu mrs
in Ihtomncdical Engtinu r-ing. 19(1): 145 (1991).
.11. Stvt z, Mart in Robert. lb yf(-Dimacasinnal .ihn'ial Imnaye A1nalys., L,ilnq Local
Dyna rnic A lgorith ma.Sc(ion O)n aI .1fhtiph -Intruction. Jbinltiph- I)ata .4icri-
thctiit. 1
P ill) (liseritation., Ill!). (Onil)Ier Scienice aidiVl Liigiiiecriiig akt thle
1Iniversit v of fi chigaii. 1989).
12. Sungiioff. Alexander kind Donald Gieeliberg. "( oliupiier Geiieraied Imlages for
MledicalI Appl)1ica tions.- (Onp t( Graphic,~. 12(l~): 196 202 (1978).
-13. Tiede. Iif. et al. liiOwigat ion of Medical 31).leiidering AlIgorithmiiis. IL
('ompuhr Owhi. ndaI,1 Apii
l)Jlliaon. 52 6~2 (May 1990).
41. Udiipa. .iayarain 1T 'lntei active Segmental ion and Houidarv Surface Fbi'-
nial ion for 3-I) ligtal Imiaes." ('onmpnh ' Graphc.. and1( lmai Ihrcsxiny.
/N(3J)11 3235 (NlarrT 1982).
45. I'dmipa. .Jayarain 1I. a md ( 4dir TU IInnani. edit or,. .)'/) bmimng /in .11f dicin(
Bosiom: ('lW P~re", 1991.
R: ViSible olne Ren"der
46. Upsoni. Craig juid Michael Keeler. \'- I -lTF
"...
(oIputer Graphics., -'2(/,):59-6-1 (August 1988).
.17. Upson, Craig, et al. "The Application \'iualization Sytem \ Comiputational
Environment for Scientific Visualization." IEEE Comp(ur Graphics and Appli-
cation.s. 30-41 (July 1989).
4S. \\atson. G. S. "Smoothing and Interpolatio) by Klriging and with Splines.'
M1ath(malical Grology. 16(6):601-15 (198-1).
49. \Vatt., Alan. Fundamentals of 7hre-Dmcnsioltl ('oinpuh r Graphics.
Addison-W-Vesley Publishing Company. 19S9.
50. \W\iihelms, .Jane and Allen Van Gelder. Topological (on.,idraoo.5 iM Isosurfa((
Gcnration. ?lechnical Report. UniversitY of (alifornia. Santa Cruz. CA. April
1990.
51. \Vilheins. Jane and Allen Van Gelder, "A Coherent 1Pro e,ion Approach for
I)irect Volume Rendering," Computer (Giaphic., 25(/):275--284 (.Julv 1991).
52. W yvil!. G., C. McPheeters and B. WVyvill. "Data Structures for Soft Objects."
1hr Iisual Computfr. 2(4):227-234 (August 1986).